article
stringlengths 0
745k
| abstract
stringlengths 0
92.7k
|
|---|---|
cu@xmath0o ( cuprous oxide ) is a p - type semiconductor that has recently attracted the attention of experimental and theoretical groups due to its potential for application in metal base transistors @xcite , spintronic @xcite , photocathode for water splitting @xcite , electrochemical supercapacitors @xcite and catalysts @xcite , and for light harvesting @xcite .
the deposition of cu@xmath0o layers has been achieved by different techniques , such as pulsed laser deposition @xcite , magnetron sputtering @xcite , copper oxidation @xcite , radical oxidation @xcite , and electrodeposition @xcite .
electrodeposition is a very versatile technique , allowing to modify many structural , optical , and electrical properties by the control of the deposition parameters @xcite .
moreover , electrodeposition appears as a cost effective method to the preparation of metal and semiconductor thin films for device applications . when compared with physical deposition processes , such as sputtering , electrodeposition has the particular feature of diffusive mass transport of species from the electrolyte to the growing surface @xcite .
a comprehensive understanding of the electrochemical growth mechanisms , taking into account the mass transport at the electrode surface , plays a vital role to the production of films with the desired properties .
one example is the technological requirement for mesoscopic layers to be used in photocathode applications , in which the grain shape and the exposed crystalline faces need to be controlled to improve stability and efficiency @xcite . on the other hand , the study of surface topography of deposited layers helps to discriminate the basic mechanisms of growth dynamics from the particular features of each set of physico - chemical deposition conditions @xcite
. a crucial condition may be the substrate , which affects the early stages of island formation and growth and , possibly , the subsequent film dynamics .
bearing in mind the above mentioned arguments , the present work aims to investigate the structure of electrochemically grown cu@xmath0o films on top of a semiconducting and monocrystalline substrate and on top of a metallic and polycrystalline substrate , viz .
n - si(100 ) and ni evaporated on n - si(100 ) .
recently , our group showed that these two substrates are suitable for morphological , structural and optical studies on cu@xmath0o @xcite .
moreover , the cu@xmath0o / ni structure may be employed on photochemical @xcite and spin transport @xcite investigations .
raman spectroscopy , x - ray diffraction ( xrd ) , and transmission electron microscopy ( tem ) measurements are carried out to characterize film composition and structure .
atomic force microscopy ( afm ) images are used to study kinetic roughening of the film surface , thus helping to disclose the main mechanisms of the growth dynamics .
films directly grown on the semiconductor substrate have a v shape columnar structure , preferential growth in @xmath2 $ ] direction , and show anomalous roughening , while those grown on the ni layer present an u shape columnar structure , change the dominant crystallographic orientation during the growth , and have normal roughening .
a deeper analysis of surface morphology shows that the dominant mechanism in the cu@xmath0o growth is surface diffusion of adsorbed molecules , with uniform incoming flux .
step edge energy barriers explain the anomaly of the former films , connecting this feature to the grain shape .
the universal coarse - grained growth dynamics is similar to what is observed in many vapor deposited films , despite the relevance of electrochemical conditions to local reaction and adsorption processes . on the other hand , the interaction between the film and the substrate , which is frequently restricted to island nucleation and growth , here
is shown to have drastic consequences to the structure of the thickest cu@xmath0o films .
the rest of this work is organized as follows . in sec .
[ secexper ] , the experimental procedure for growth and characterization of the cu@xmath0o films is described . in sec .
[ secresults ] , the experimental results are presented and analyzed in the light of kinetic roughening concepts .
section [ secconclusion ] presents a discussion of the results and our conclusions .
cu@xmath0o films were deposited on ( 100 ) monocrystalline n - type silicon ( resistivities in the range of 6 - 9 @xmath3 cm ) without and with a cap layer of 50 nm of evaporated ni . before cu@xmath0o electrodeposition or ni deposition , the silicon substrates were immersed in hf 5% for 20s to remove silicon native oxide from the surface .
the roughnesses of the n - si substrate and of the ni layer surface are respectively @xmath4 nm and @xmath5 nm . ni thin films deposited on si(100 ) had preferential growth in the @xmath6 $ ] direction , which was checked by xrd measurements .
the electrochemical experiments were conducted in a conventional three electrode cell connected to a potentiostat autolab pgstat30 .
the counter electrode was a pt foil and the reference electrode was a saturated calomel electrode ( sce ) .
the electrolyte , prepared from analytical grade reagents and deionized water ( resistivity of 18 m @xmath3 cm ) , contains @xmath7 m cuso@xmath8 and @xmath9 m lactic acid , with the ph adjusted to @xmath10 by adding a solution of @xmath11 m naoh @xcite .
the deposits were obtained via potentiostatic experiments .
samples were deposited at 25 for a deposition potential of @xmath12 v _ vs _ sce , where the only varied parameter was the thickness @xmath13 of the samples for the 2 different substrates , n - si(100 ) and ni / n - si(100 ) .
the efficiency of the cu@xmath0o electrodeposition process was checked by rutherford backscattering in a previous work and values of 90% were obtained @xcite .
the samples were characterized by various techniques including raman spectroscopy ( invia , renishaw ) , xrd ( xpert , philips ) , tem ( jem-2100 , jeol ) and afm ( pico - spm , molecular imaging corporation ) .
raman spectra were obtained from @xmath14 nm wavelength excitation ( argon ion laser ) .
the afm images were acquired in contact mode in air at room temperature and the values of the root mean - square deviation of the local surface height , @xmath15 , are representative of at least three images of different regions of the surface .
three different regions for sample were measured with scan sizes of @xmath16 , @xmath17 and @xmath18 for each region and number of pixels per line ranging from 256 to 1024 , respectively . from these measurements
we have concluded that for @xmath19 image size the @xmath15 reaches the saturation value .
in addition , before starting measurements on cu@xmath0o samples , the performance of the afm tip was checked by imaging au thin films with known surface .
the copper oxide phase formed in the electrodeposited thin films was analyzed by raman spectroscopy and xrd .
the measurements were done on layers with thickness between 150 and 1500 nm deposited on n - si(100 ) and ni / n - si(100 ) .
raman spectroscopy results of films with thickness @xmath20 nm are shown in fig .
the predominance of cu@xmath0o phase is evident .
contributions of cu@xmath8o@xmath21 and possibly cuo phases are also observed .
the peaks in fig .
[ fig1]a were addressed to the respective phase and raman vibration mode based on ref .
despite the observation of cu@xmath8o@xmath21 and cuo raman peaks , these phases are not present in xrd bragg - brentano patterns of the same films , shown in fig .
[ fig1]b , probably due to the higher surface sensitivity of raman measurements . cu@xmath8o@xmath21 and cuo phases are likely to be formed by oxidation of cu@xmath0o top layers
. since the layers are mostly of cu@xmath0o , this work will be concerned to the description of the electrochemical growth of this phase . o films of 1250 nm electrodeposited on n - si and ni substrates . for each raman peak
are indicated the possible related crystalline phases and , between parentheses , the raman vibrational mode .
xrd characterization was carried out in bragg - brentano configuration .
the inset in b ) shows the cu@xmath0o portion growing in @xmath2 $ ] and @xmath6 $ ] directions as a function of the film thickness.,width=321 ] fig .
[ fig1]b also shows the xrd patterns of cu@xmath0o films with 1250 nm of thickness deposited on ni / n - si(100 ) .
the peaks are also in the expected positions for the cu@xmath0o crystal .
cu@xmath0o / n - si(100 ) samples display strong preferential growth in @xmath2 $ ] direction , following the orientation of the substrate .
on the other hand , cu@xmath0o film grown on top of ni / n - si(100 ) substrate is composed by grains oriented in @xmath2 $ ] , @xmath22 $ ] , @xmath6 $ ] , and @xmath23 $ ] directions .
the evolution of cu@xmath0o growth on ni / n - si(100 ) is explained by ref . @xcite
as follows : ( _ i _ ) the cu@xmath0o films show initial @xmath6 $ ] growth due to a better coupling of the cu@xmath0o(111)/ni(111 ) interface compared to the cu@xmath0o(100)/ni(111 ) one ; ( _ ii _ ) at electrolyte ph of @xmath10 the @xmath2 $ ] growth is favored , the initial @xmath6 $ ] growth starts to be suppressed by the @xmath2 $ ] one and at a specific thickness a major part of the cu@xmath0o deposit will be composed by @xmath2 $ ] grains .
these growth steps are confirmed by the inset of fig .
[ fig1]b , which shows the fraction of the material growing in @xmath2 $ ] and @xmath6 $ ] directions as a function of film thickness for both substrates .
while cu@xmath0o on n - si(100 ) is , independently of film thickness , oriented in @xmath2 $ ] direction , cu@xmath0o on ni / n - si(100 ) grows up to 750 nm with @xmath24% of its grains oriented in @xmath6 $ ] direction , but only @xmath25% remain with this orientation in larger thicknesses .
the initial growth of cu@xmath0o on ni(111 ) is thermodynamically controlled by the cu@xmath0o(111)/ni(111 ) coupling , likewise as previously reported for cu@xmath0o(111)/au(111 ) and cu@xmath0o(100)/au(100 ) interfaces @xcite .
however , the crystallographic orientation transition at 750 nm is related to the oxygen concentration in electrolyte at ph @xmath10 , which favors the @xmath2 $ ] growth @xcite .
the structure of cu@xmath0o films was checked by cross section tem images .
figure [ fig2]a shows an image obtained from electrodeposited cu@xmath0o film on n - si(100 ) .
a high density columnar microstructure without voids is observed .
these columns have a v shape as stressed by discontinuous red lines , exhibiting a grain width increase as a function of film thickness . on the other hand , as seen in fig .
[ fig2]b , cu@xmath0o films deposited on the ni layer under same conditions ( solution , temperature and deposition potential ) displayed a columnar microstructure with u shape , as indicated in fig .
[ fig2]b , and grain width roughly constant as a function of growth time . o thin films electrodeposited respectively on a ) n - si and b ) ni
different layers in each image are labeled as follows : 1 cu@xmath0o film , 2 n - si substrate , 3 ni film ( substrate ) and 4 the glue used on sample preparation . in a ) and b ) discontinuous red lines highlight the v shape and u shape of the cu@xmath0o columns , respectively.,width=321 ] figure [ fig3 ] shows @xmath26 afm images of the films with thicknesses of 250 and 1500 nm .
a common feature in both substrates is that the film surface presents a granular aspect . in films deposited on n - si
, there is a remarkable increase of the grain size from the thinnest to the thickest film .
the columnar structure with v shape shown in tem images and this increase in grain size suggest a coarsening process , with part of the initial columns growing and enlarging up to 1500 nm at the expenses of neighboring ones . on the other hand ,
a mild increase of grain size is observed in the films grown on ni / n - si , which is in agreement with tem images showing columns in u shape .
afm images of the surface morphology of cu@xmath0o layers electrodeposited on : a ) , b ) n - si ; and c ) , d ) ni / n - si .
upper row is for thickness of 250 nm , while lower row is for 1500 nm .
bright areas are elevations and dark ones are deep regions.,width=321 ] the surface roughness @xmath27 is the most used quantity to characterize height fluctuations . for a given surface under a process of kinetic roughening , @xmath27 depends on the lengthscale in which fluctuations are measured .
moreover , given a scanning box with lateral size @xmath28 , the fluctuation in film height inside the box also depends on the box position .
for these reasons , the box roughness is defined as the root - mean - square ( rms ) fluctuation of the film height ( @xmath29 ) inside a box at a given position , i.e , @xmath30 .
the average local roughness @xmath31 of a film of thickness @xmath13 in the lengthscale @xmath28 is the average of the box roughness @xmath32 among all possible box positions .
@xmath15 is hereafter simply called roughness . using the heights from afm images ,
@xmath15 is calculated for cu@xmath0o layers grown on both substrates .
the results are shown in figs .
[ fig4]a and [ fig4]b for a range of thicknesses @xmath13 varying from 25 to 1500 nm .
the structural changes induced by substrates
n - si and ni / n - si are reflected in topographic differences in the growing surface of the films , confirming the visual inspection of afm images . as a function of the box size @xmath28 measured from afm images of a series of electrodeposited cu@xmath0o films for thickness @xmath13 from 25 to 1500 nm .
in a ) and b ) the substrate is n - si and ni / n - si , respectively.,width=321 ] the roughness @xmath15 of films grown on ni / n - si show normal dynamic scaling because it is approximately time - independent for small box size @xmath28 and saturates at values increasing with the thickness @xmath13 .
deviations for the thinner samples ( 25 to 75 nm ) are characteristic of the initial island growth and coalescence , indicating that an extended film is formed between 75 nm and 150 nm of thickness . for this reason , only results for @xmath33 nm will be analyzed below .
on the other hand , the oxide layers grown on n - si have anomalous scaling , since the roughness in small boxes shift to larger values with increasing thicknesses @xcite .
the presence of a single dominant orientation during the growth ( shown by xrd ) and the coarsening of growing columns from the smallest to the largest thicknesses ( suggested by tem and afm images ) indicate that the scaling anomaly is an intrinsic feature of the surface dynamics in this system . the data presented in fig .
[ fig4 ] for the thickest films with @xmath34 nm were fit to the relation @xmath35 .
it gives estimates of the local roughness exponents @xmath36 and @xmath37 in n - si and ni / n - si substrates , respectively .
these values are consistent with the interpretation of granular effects in refs . @xcite and @xcite , which predict an exponent near 1 if the maximal box size is smaller than the grain size .
inspection of the afm images ( fig .
[ fig3 ] ) supports this hypothesis .
these estimates of @xmath38 must not be interpreted as true exponents of dynamic scaling , but as exponents representing the grain shapes @xcite .
the local roughness @xmath39 is measured for a fixed small box size @xmath40 nm .
it is shown in fig . [ fig5]a as a function of the thickness @xmath13 . for films deposited on ni
, @xmath41 fluctuates around a constant value as the thickness increases , from small values to the largest ones .
this is consistent with normal scaling . on the other hand , for films deposited on n - si
, @xmath41 shows a power law increase from small thicknesses ( @xmath42 nm ) to the largest ones ( @xmath43 nm ) .
this is the main evidence of anomalous scaling in these films .
the fit of the data in fig .
[ fig5]a by the scaling relation @xmath44 @xcite gives an anomaly exponent @xmath45 ( @xmath46 is also called local slope exponent in theoretical works and local growth exponent @xmath47 in experimental works @xcite ) .
the saturation values of the roughness in fig .
[ fig4 ] give the global roughness @xmath48 , which characterize fluctuations of the whole surface at a given thickness . in fig .
[ fig5]b , @xmath48 is shown as a function of @xmath13 for both films .
the fits to the scaling relation @xmath49 for @xmath33 give the growth exponents @xmath50 and @xmath51 for films deposited on n - si and ni / n - si substrates , respectively .
since the growth on ni / n - si substrates has normal scaling , we search for a model of local stochastic growth equation , which may disclose the main mechanisms of the surface dynamics @xcite .
the above estimate of @xmath52 is larger than the values provided by the equations of second and fourth orders , which are in the range @xmath53 $ ] @xcite .
however , fig .
[ fig5]b shows that a slope close to @xmath54 can not be discarded for large thicknesses .
this is close to the exponents of the kardar - parisi - zhang ( kpz ) class @xcite , @xmath55 @xcite , and of the mullins - herring ( mh ) class @xcite , @xmath54 .
information on surface correlations can be obtained from the spatial correlation function @xmath56 at time @xmath57 , with @xmath58 .
the rescaled functions @xmath59 of the oxide films respectively grown on n - si and ni / n - si substrates are shown in figs .
[ fig5]c and [ fig5]e as a function of the distance @xmath28 , for several thicknesses .
a reliable estimate of the correlation length @xmath60 is the minima of @xmath59 @xcite , which is shown in figs .
[ fig5]d and [ fig5]f as a function of the thickness @xmath13 .
the estimates of @xmath60 are close to those obtained by intersection of straight line fits of the roughness in the growth and saturation regimes , which is a usual method in the analysis of experimental data . ref .
@xcite shows that this method leads to estimates of @xmath60 of the same order of the average grain size estimated by inspection of microscopy images @xcite .
however , an accurate calculation of the average grain size is more complicated @xcite and not performed here .
the scaling @xmath61 is expected for large thicknesses @xcite .
linear fits of the data for @xmath62 nm give estimates @xmath63 and @xmath64 ( dynamic exponents @xmath65 and @xmath66 ) for growth on n - si and ni / n - si substrates , respectively .
these exponents are very different from the kpz value @xmath67 @xcite . on the other hand , the figs .
[ fig5]d and [ fig5]f show that the slopes of the @xmath68 plots for the largest thicknesses are close to the mh value @xmath69 ( the corresponding slope of kpz scaling @xmath70 is also shown for comparison ) .
the mh equation is a model for growth dominated by surface diffusion of adsorbed species @xcite : @xmath71 where @xmath72 is the interface height at substrate position @xmath73 and time @xmath57 , @xmath74 is a constant , and @xmath75 is a gaussian ( nonconservative ) noise with covariance amplitude @xmath76 @xcite .
a constant external flux was omitted from eq .
( [ eqmh ] ) , which corresponds to a spatially uniform and time - independent adsorption rate .
thus , the local growth rate of the mh model is not affected by diffusion in solution , for instance due to the presence of diffusive layers , shadowing effects etc . on the other hand ,
the kpz equation is a model for interface growth dominated by surface tension and by a nonlinear effect of the local slope @xcite : @xmath77 where @xmath78 and @xmath79 are constants .
distinguishing these possibilities is an important step to understand the microscopic growth dynamics of the oxide films .
the kpz model also assumes spatially uniform and time - independent adsorption rate , with no effect of diffusion in solution to the local growth rate .
we now turn to the analysis of distributions of heights and of local quantities , which was recently shown to be a more powerful tool to the study of kinetic roughening of thin films @xcite .
the height distribution @xmath80 is defined as the probability density of the height @xmath29 , so that @xmath81 gives the probability of finding a height in the range @xmath82 $ ] .
figures [ fig6]a and [ fig6]c , and [ fig6]b and [ fig6]d show the scaled height distributions of the oxide films with large thicknesses , respectively grown on n - si and ni / n - si substrates .
they are compared with distributions of the kpz and mh classes in the growth regimes .
the kpz distribution was numerically calculated in refs . @xcite and @xcite . for the mh class ,
the distribution is gaussian , which was confirmed by direct integration of the mh equation [ eq . [ eqmh ] ] and by simulation of the large curvature model ( lcm ) @xcite . the experimental data in figs .
[ fig6]a , [ fig6]b , [ fig6]c and [ fig6]d are closer to the gaussian / mh curve .
the symmetric shape clearly contrasts with the asymmetry of the kpz curve .
indeed , the skewness @xmath83 and kurtosis @xmath84 of the @xmath80 distributions , shown in figs .
[ fig6]e and [ fig6]f , drops to zero as the thickness increases ( consistently with gaussian / mh ) , while the asymptotic kpz values are @xmath85 and @xmath86 @xcite .
another important conclusion emerges from figs .
[ fig6]a and [ fig6]b : since the scaled height distributions are the same for films deposited on n - si and on ni / n - si , the main physico - chemical processes responsible for their roughening are the same .
the distribution @xmath87 of the squared box roughness @xmath88 is defined so that @xmath89 is the probability that the squared roughness in a given box is in the range @xmath90 $ ] @xcite .
figures [ fig7]a and [ fig7]c compares the scaled roughness distributions of the oxide films grown on both substrates and that of the mh class in the growth regime ( @xmath91 is the rms fluctuation of @xmath92 ) .
the size of the box in which @xmath92 is measured ranges from 100 nm to 300 nm .
the mh curve is calculated by integration of the mh equation in the growth regime . in figs .
[ fig7]a and [ fig7]c , the collapse of the scaled distributions for films grown in different substrates is clear , which is additional support to the proposal that their roughening is governed by the same physico - chemical processes ( despite the very different structures of those films ) .
good agreement with the mh curve reinforces the proposal that diffusion of adsorbed species is the main mechanism of the surface dynamics .
we also analyzed the distribution @xmath93 of the maximal relative height @xmath94 in the growth regime .
here , @xmath95 and @xmath96 are , respectively , the maximal and the average heights measured inside each box position . the distributions for both films are shown in figs .
[ fig7]b and [ fig7]d , showing good data collapse among them and with the mh curve . following the procedure described in ref .
@xcite , we also calculated the local surface slope @xmath97 using the afm images . the local slope images corresponding to the height images shown in fig .
[ fig3 ] are displayed in fig .
they clearly show that larger values for @xmath97 are obtained in intergranular regions , while small @xmath97 are related to flat areas near the top of cu@xmath0o grains .
the distributions of local slopes are shown in figs .
[ fig8]e and [ fig8]f for cu@xmath0o / n - si and cu@xmath0o / ni films , respectively .
the position @xmath98 of the peaks of those distributions are displayed in fig .
[ fig8]g as a function of thickness
. for films grown on n - si , the distribution significantly changes as the thickness increases , with an increase in @xmath98 .
this is typical of anomalous scaling @xcite and is qualitatively consistent with the local roughness increase shown in fig .
[ fig5]a . on the other hand , for films grown on ni / n - si , average local slopes decrease with the thickness , saturating in the thicker films .
this is consistent with normal scaling .
comparison of local slope distributions with those of theoretical models is not possible because there are significant thickness effects and the present concept of local slope differs from the definitions in integrated growth equations or in lattice models .
the mh scaling was formerly observed in electrochemically deposited prussian blue films @xcite . in that case ,
the anomalous roughening was generated by time increasing adsorption rates .
the exact solution of the mh equation @xcite allows an extension of the interpretation of the experimental data , particularly those of films grown on n - si substrates . the mh roughness @xmath15 is expected to scale similarly to the height - height correlation function ; in terms of box size @xmath28 and time @xmath57 we have @xcite @xmath99 where @xmath100 is a scaling function [ @xmath101 for small @xmath102 and @xmath103 recall that @xmath13 is proportional to the time @xmath57 . the long time ( large thickness ) scaling of the correlation length gives exponent @xmath104 near the mh value ( @xmath105 ) .
thus , eq . [ eqximh ] indicates that @xmath74 is approximately time - independent .
the coefficient @xmath74 in eq .
[ eqmh ] represents the relation between surface diffusion coefficients and local surface geometry of a given material .
this relation is not expected to change in the course of the ( constant temperature ) deposition . for @xmath106 , eq .
[ eqwmh ] suggests that the local roughness @xmath107 ( measured for fixed @xmath108 nm ; sec .
[ resdynamic ] ) scales as @xmath109 .
figure [ fig9 ] shows @xmath110 versus @xmath13 for both films . the approximately constant value of @xmath111 for the films grown on ni / n - si indicates that @xmath76 is also time - independent in that case .
however , the scaling in films grown on n - si and the time - independence of @xmath74 gives @xmath112 .
since @xmath76 is the amplitude of time and spatial fluctuations of deposition rate , this means that those fluctuations increase in time in the films grown on n - si .
this leads to the anomalous scaling in those films . as a function of the thickness @xmath13 for films grown on n - si ( triangles ) and on ni / n - si ( squares )
the solid line , with slope @xmath113 , is a least squares fit of the large thickness data for films grown on n - si.,width=302 ] the above increase of @xmath76 is sufficiently slow , thus the original solution of the mh equation can be consistently used with the time - dependent form of that parameter @xcite .
this feature was already illustrated in other continuous and lattice models with time dependent couplings @xcite showing anomalous scaling .
the scaling of distributions of local heights , maximal heights , and roughness provide strong evidence that the roughening in electrodeposited cu@xmath0o films is governed by the mh equation ( [ eqmh ] ) .
the noise amplitude increases in time in the equation associated to the films grown on n - si substrates , corresponding to the scaling anomaly .
this analysis clearly advance over the calculation of scaling exponents , which however are also compatible with the mh ones .
the mh equation was proposed for roughening in molecular beam epitaxy @xcite , in which a crystalline film slowly grows by random adsorption of atoms / molecules from vapor and surface adatoms diffuse and preferentially bind to high coordination ( low energy ) sites .
thus , in the electrochemical deposition of cu@xmath0o films , those results indicate that this interplay of surface diffusion and deposition noise is also the main ingredient for roughening under the present growth conditions .
the mh equation considers that adsorption rates are spatially uniform and do not depend on time .
this suggests that electrochemical conditions do not determine the main dynamic mechanisms for roughening , although they probably affect the values of the parameters in the associated mh equation .
since the same electrochemical conditions were used for growth of films on both substrates , we are led to the conclusion that their different structures are consequence of the substrate - film interaction .
the different interactions in different substrates determine the island nucleation and growth and has consequences in the thickest films , despite the universal mh dynamics in systems with diffusion - dominated dynamics , molecules at a surface grain randomly move , with rates depending on local energy barriers . at first approximation ,
those barriers depend on coordination numbers and increase at step edges . for an adatom or ad - molecule to go down a step
, the additional activation energy , known as ehrlich - schwoebel ( es ) barrier @xcite , reduces the downward flux near the step edge , favoring nucleation of new atomic terraces over the ones previously formed @xcite .
the effect of es barriers is enhanced at grain boundaries , constraining the mass flux between the grains if height differences are large .
these features are observed in films of several materials grown by vapor methods .
the above results strongly suggest to extend this interpretation to cu@xmath0o electrodeposition .
figure [ fig3]a shows that films grown on n - si substrates have an initially small average grain size , but with a broad size distribution .
this is confirmed quantitatively by the small values of the correlation length in fig .
[ fig5]d ( @xmath114 nm in @xmath115 nm thick films ) .
thus , the islands nucleating at that substrate also had large size fluctuations
. this may be a consequence of large spatial fluctuations of the adsorption rate during the first stages of the growth .
when an extended film is formed , the local adsorption rates may become uniform , particularly if the dominant crystallographic orientation does not change , as shown in sec .
[ rescharact ] ; see also inset of fig .
however , larger grains have larger numbers of terraces and steps , consequently the mass flux from their tops to their boundaries is small . with restricted mass exchange in grain boundaries , the large grains increase in width and
cover the smaller ones . from a coarse grained point of view
, the fluctuations in the adsorption rate continue to increase , which qualitatively explains the time increase of the parameter @xmath76 in the corresponding hydrodynamic model ( the mh equation ) .
figure [ fig8]e shows that the shape of the slope distribution is the same while its average value moves to the right during the film growth .
this means that local slopes increase inside the grains and at their boundaries , consistently with the above interpretation .
on the other hand , fig .
[ fig3]c shows that films grown on ni / n - si substrates have larger initial grains with a much narrower size distribution ( see also insets of figs .
[ fig5]d and [ fig5]f ) .
this is related to small fluctuations in local adsorption rates , a feature that is maintained during the growth and leads to normal roughening .
afm images also reveal that the grain top surfaces are similar in the films grown in both substrates , thus the local ( microscopic ) adsorption rates at those surfaces are expected to be the same . in growth on n - si substrates , the difference in the net growth rate of individual grains have to be explained by the restricted mass flux across grain boundaries , not by intra - grain features .
if some grains with large slopes are initially formed , they may grow faster at the expenses of the neighboring ones .
this establishes a connection between the v - shaped grains and the anomalous roughening .
unusual scaling features are frequently found in diffusion - dominated growth in the presence of step energy barriers @xcite .
for instance , a small barrier in fe / fe(100 ) epitaxy leads to mound formation at the film surface , with skewed height distribution and @xmath116 @xcite , which are characteristics of the non - linear molecular beam epitaxy equation @xcite .
moreover , mound steepening in ag / ag(100 ) epitaxy at room temperature occurs with an effective exponent @xmath52 much larger than @xmath117 in thicknesses ranging from 100 to 1000 monolayers @xcite ( @xmath118 is a feature of completely uncorrelated growth @xcite ) .
we conclude that , although control of electrochemical conditions is essential to enable growth of cu@xmath0o films and to determine the crystallographic orientation , our results show that large scale surface features are determined by physical properties of the material and the substrate , with consequences on the internal film structure .
roughening is governed by the interplay of deposition and ad - species surface diffusion , similarly to vapor deposition , with no significant influence of inhomogeneous mass flux from the solution ( if it exists ) .
however , substrate - film interactions have a strong effect on island nucleation and growth , thus the structure of the tickest films is drastically affected by these initial conditions , despite the universal ( mh ) roughening dynamics .
the authors acknowledge financial support from capes , cnpq ( namitec and nanoinstrumentao ) , fapesc , fapemig and faperj ( brazilian agencies ) .
lcme , labmat and ldrx laboratories of the universidade federal de santa catarina for the use of the tem , raman spectroscopy and xrd facilities . | cuprous oxide ( cu@xmath0o ) films from 25 nm to 1500 nm were electrodeposited on n - si(100 ) and ni / n - si(100 ) substrates from aqueous solution at room temperature .
x - ray diffraction and transmission electron microscopy imaging show that the cu@xmath0o structure and morphology is strongly affected by the substrate choice , with v shape and u shape columnar growth on n - si(100 ) and ni / n - si(100 ) , respectively . atomic force microscopy reveals the presence of rounded grains at the surface in both cases .
anomalous and normal roughening are observed in films grown on n - si and ni , respectively , but estimates of scaling exponents are not conclusive . on the other hand ,
the distributions of local heights , roughness , and extremal heights show good agreement with those of the fourth order linear stochastic equation of mullins and herring ( mh ) .
thus , surface dynamics in both systems is dominated by diffusion of adsorbed molecules , with no large scale effect of possible inhomogeneities in mass flux from the solution or in reaction and adsorption rates . in growth on n - si substrates ,
the noise amplitude of the mh equation increases in time as @xmath1 , while the coefficient of the curvature - related term is time - independent .
step edge energy barriers restrict the mass flux across grain boundaries , thus a broad size distribution of initial grains leads to coarsening of the larger ones .
this explains their v shape in the thickest films and establishes a connection with the anomalous roughening .
these effects are reduced in films grown on ni / n - si , which initially have much larger grains with narrower size distributions and , consequently , smaller fluctuations in coarse grained growth rates .
thus , despite the relevance of electrochemical conditions for cu@xmath0o films to grow and their influence on crystallographic orientation , large scale surface features are determined by physical properties of the material and its interactions with the substrate , with a universal microscopic dynamics similar to vapor deposition . |
orbifolds or v - manifolds were first introduced by satake @xcite , and arise naturally in many ways .
for example , the orbit space of any proper action by a ( discrete ) group on a manifold has the structure of an orbifold ; this applies in particular to moduli spaces .
furthermore , the orbit space of any almost free action by a compact lie group has the structure of an orbifold , as does the leaf space of any foliation with compact leaves and finite holonomy .
examples of orbifolds are discussed in @xcite and many others . for an orbifold @xmath0
, one can define in a natural way a cohomology theory with coefficients in any local system on @xmath0 .
this cohomology is not an invariant of the underlying ( ` coarse ' ) space , but of the finer orbifold structure .
if the orbifold is given as the orbit space @xmath2 of a group action as above , this cohomology is the equivariant sheaf cohomology of the group action .
it agrees with the ( ordinary ) cohomology of the borel construction @xmath3 .
this cohomology is the most natural one for orbifolds .
it fits in well with the notion of fundamental group described in @xcite , by the familiar ` hurewicz formula ' @xmath4 ( where @xmath5 is any abelian group ) .
the purpose of this paper is to give a _ simplicial _ description of these cohomology groups , suitable for calculations .
more precisely , using triangulations of singular spaces @xcite , we will associate to any orbifold @xmath0 , presented by an orbifold atlas as in @xcite , a simplicial set @xmath1 .
the construction of @xmath1 uses the simplices in a triangulation of the coarse underlying space @xmath6 of @xmath0 , as well as all the local isotropy groups
. the construction will have the following property .
[ [ section ] ] * theorem * _ for any local system of coefficients @xmath7 on the orbifold @xmath0 , there is a canonically associated local system @xmath5 on the simplicial set @xmath1 , for which there is a natural isomorphism @xmath8 . _
[ [ section-1 ] ] after reviewing some preliminary definitions , we will present our construction in section [ simpcom ] of this paper .
the proof of the theorem will be based on the fact that the simplicial set @xmath1 associated to an orbifold @xmath0 is also closely related to the representation of @xmath0 by a groupoid @xmath9 suggested in @xcite .
in fact , our proof shows that @xmath1 has the same homotopy type as the classifying space of @xmath9 . before giving the proof in section [ seven ] , we will present an example of calculations based on the simplicial construction for the ` teardrop ' orbifold .
we believe that much more work should be done in this direction .
in fact , the explicit description of the simplicial set in terms of an atlas for the orbifold , and the resulting description of the cohomology groups by generators and relations , makes it suitable for computer assisted calculations .
in this section we briefly review the basic definitions concerning orbifolds , or v - manifolds in the terminology of satake ( see @xcite ) .
let @xmath6 be a paracompact hausdorff space .
an _ orbifold chart _ on @xmath6 is given by a connected open subset @xmath10 for some integer @xmath11 , a finite group @xmath12 of @xmath13-automorphisms of @xmath14 , and a map @xmath15 , such that @xmath16 is @xmath12-invariant ( @xmath17 for all @xmath18 ) and induces a homeomorphism of @xmath19 onto the open subset @xmath20 . an _
embedding _ @xmath21 between two such charts is a smooth embedding @xmath22 with @xmath23
. an _ orbifold atlas _ on @xmath6 is a family @xmath24 of such charts , which cover @xmath6 and are locally compatible in the following sense : given any two charts @xmath25 for @xmath26 and @xmath27 for @xmath28 , and a point @xmath29 , there exists an open neighborhood @xmath30 of @xmath31 and a chart @xmath32 for @xmath33 such that there are embeddings @xmath34 and @xmath35 .
two such atlases are said to be equivalent if they have a common refinement .
an _ orbifold _ ( of dimension @xmath36 ) is such a space @xmath6 with an equivalence class of atlases @xmath37 .
we will generally write @xmath38 for the orbifold @xmath0 represented by the space @xmath6 and a chosen atlas @xmath37 .
[ two ] for two embeddings @xmath39 between charts , there exists a unique @xmath40 such that @xmath41 .
in particular , since each @xmath18 can be viewed as an embedding of @xmath42 into itself , there exists for the two embeddings @xmath43 and @xmath44 a unique @xmath40 with @xmath45 .
this @xmath46 will be denoted by @xmath47 . in this way
, every embedding @xmath43 also induces an injective group homomorphism , ( again denoted ) @xmath48 , with defining equation @xmath49 furthermore , if @xmath40 is such that @xmath50 , then @xmath46 belongs to the image of this group homomorphism @xmath48 , and hence @xmath51 .
( this is proved in @xcite for the codimension 2 case , and in @xcite for the general case . ) by the differentiable slice theorem for smooth group actions @xcite , any orbifold of dimension @xmath36 has an atlas consisting of ` linear ' charts , i.e. charts of the form @xmath52 where @xmath12 is a finite group of linear transformations and @xmath53 is an open ball in @xmath54 .
if @xmath25 and @xmath27 are two charts for the same orbifold structure on @xmath6 , and @xmath53 is simply connected , then there exists an embedding @xmath55 whenever @xmath56 , i.e. when @xmath57 ( see @xcite , footnote 2 ) .
_ in this paper we will take all charts to be simply connected .
_ [ contract ] if @xmath25 is a chart , and @xmath58 is a connected ( and simply connected ) open subset of @xmath59 , then @xmath58 inherits a chart structure from @xmath60 in the following way : let @xmath61 be a connected component of @xmath62 , and let @xmath63 .
then @xmath64 is a chart , which embeds into @xmath42 , and hence defines the same orbifold structure on points in @xmath58 .
we list some well - known examples , see e.g. @xcite .
if a discrete group @xmath65 acts smoothly and properly on a manifold @xmath66 , the orbit space @xmath67 has a natural orbifold structure .
examples are weighted projective space and moduli spaces .
if a compact lie group @xmath68 acts smoothly on a manifold @xmath66 and the isotropy group @xmath69 at each point @xmath70 is finite and acts faithfully on a slice @xmath71 through @xmath31 , then the orbit space @xmath72 has a natural orbifold structure .
moreover any orbifold can be represented this way . if @xmath66 is a manifold equipped with a foliation @xmath73 of codimension @xmath36 , with the property that each leaf is compact and the holonomy group at each point is finite , then the space of leaves @xmath74 has the natural structure of an @xmath36-dimensional orbifold .
again , any orbifold can be represented in this way .
let @xmath38 be an orbifold of dimension @xmath36 .
for each point @xmath75 , one can choose a linear chart @xmath42 around @xmath31 , with @xmath12 a finite subgroup of the linear group @xmath76 .
let @xmath77 be a point with @xmath78 , and @xmath79 the isotropy subgroup at @xmath80 .
up to conjugation , @xmath81 is a well defined subgroup of @xmath76 .
the space @xmath6 thus carries a well - known natural stratification , whose strata are the connected components of the sets @xmath82 where @xmath83 is any finite subgroup of @xmath76 and @xmath84 is its conjugacy class .
it is also well - known ( @xcite ) that there exists a triangulation @xmath85 of @xmath6 subordinate to this stratification ( i.e. the closures of the strata lie on subcomplexes of the triangulation @xmath85 ) . by replacing @xmath85 by a stellar subdivision , one can assume that the cover of closed simplices in @xmath85 refines the atlas @xmath37 . for a simplex in such a triangulation ,
the isotropy groups of all interior points of @xmath86 are the same , and are subgroups of the isotropy groups of the boundary points . by taking a further subdivision of @xmath85
, we may in fact assume that there is one face @xmath87 such that the isotropy is constant on @xmath88 , and possibly larger on @xmath89 .
in particular , any simplex @xmath86 will then have a vertex @xmath90 with maximal isotropy , i.e. @xmath91 for all @xmath92 .
we call such a triangulation _ adapted to _ @xmath37 . for reference we state : for any orbifold @xmath6 and any orbifold atlas @xmath37 there exists an adapted triangulation @xmath85 for @xmath38 .
let @xmath86 be a ( closed ) @xmath36-simplex in such a triangulation , which is contained in a chart @xmath60 .
( in this paper we will take all simplices to be closed . )
the next lemma describes how to lift @xmath86 to a simplex @xmath93 in @xmath53 by choosing a connected component of the inverse image @xmath94 of the interior .
[ lift ] for every connected component @xmath95 of the inverse image @xmath94 of the interior of @xmath86 , the map @xmath16 restricts to a homeomorphism on the closures @xmath96 ; in particular the triangulation of @xmath97 lifts to a triangulation of @xmath98 .
[ [ proof . ] ] proof .
+ + + + + + as is well - known ( see e.g. ( * ? ? ?
* lemma 1 ) ) , the map @xmath99 restricts to a covering projection on each stratum .
in particular , since the isotropy is constant on the interior @xmath100 , we find that @xmath101 is a disjoint sum of open simplices @xmath102 with @xmath103 .
let @xmath95 be one of these @xmath102 . by continuity , @xmath16 maps @xmath98 to @xmath97 .
this restriction is also surjective , because @xmath104 and the action of @xmath12 on @xmath53 is continuous .
since @xmath85 is subordinate to the stratification , the isotropy groups of the boundary points of @xmath86 contain the isotropy group of the interior points , and therefore @xmath16 has to be one - one on @xmath98 .
since @xmath16 is also open ( as quotient map ) , it follows that @xmath105 .
[ [ section-2 ] ] by taking further stellar subdivisions of @xmath85 , we may assume that for every simplex @xmath86 in @xmath85 , the closure of the open star @xmath106 is contained in a chart of the atlas @xmath37 . we will call a triangulation of @xmath0 with these properties a _ good _ triangulation for @xmath38 . for any orbifold @xmath6 and any orbifold atlas @xmath37
there exists a good triangulation @xmath85 for @xmath38 .
consider such a good triangulation .
let @xmath31 be a point of @xmath6 , let @xmath107 be the smallest simplex containing @xmath31 , and consider the open star neighborhood @xmath108 .
let @xmath25 be a chart in the atlas @xmath37 for which @xmath109 . for later purposes , it is useful to describe how @xmath110 lifts to a triangulation in @xmath53 .
[ stars ] the closed star @xmath110 lifts to a triangulation @xmath111 of @xmath112 with the property that every connected component of @xmath113 contains precisely one lifting @xmath80 of @xmath31 , and these components are open star neighborhoods @xmath114 for the triangulation @xmath111 : @xmath115 [ [ proof.-1 ] ] proof .
+ + + + + + according to lemma [ lift ] , for every @xmath36-simplex @xmath116 and every connected component of @xmath117 there is precisely one lifting @xmath118 of @xmath119 containing that connected component . since we could lift the whole triangulation of @xmath119 to @xmath118 , we get together with @xmath118 also all its boundary simplices .
let @xmath111 be the set of all those liftings ( for all @xmath36-simplices in @xmath110 and their boundaries ) .
it is clear that they cover @xmath120 .
remark that for every simplex @xmath121 , @xmath122 becomes a homeomorphism onto a simplex in @xmath110 .
let @xmath123 be two simplices with @xmath124
. then @xmath125 .
this last intersection is a simplex in @xmath85 , which we will denote by @xmath119 .
let @xmath126 be the two liftings of @xmath119 in the @xmath127 .
( note that @xmath128 . )
there exists an element @xmath18 such that @xmath129 .
this element @xmath130 is in the isotropy group of @xmath131 , but not in the isotropy group of the rest of @xmath119 .
since @xmath85 is subordinate to the stratification induced by the isotropy groups , it follows that @xmath131 is a subsimplex of @xmath119 , and therefore a subsimplex of @xmath132 .
we know already that @xmath133 , so @xmath134 is a simplex in @xmath111 .
now consider two liftings @xmath80 and @xmath135 of @xmath31 in @xmath53 , and their open star neighborhoods @xmath136 and @xmath137 .
suppose @xmath138 , and let @xmath139 be a point in this intersection .
thus @xmath140 where @xmath141 and @xmath142 . but @xmath143 is one - one on @xmath119 , as shown in lemma [ lift ] .
so @xmath144 and @xmath145 .
this proves the lemma .
[ goodcov ] let @xmath38 be any orbifold with atlas @xmath37 .
there exists an atlas @xmath146 for @xmath0 such that @xmath146 refines @xmath37 ; for every chart @xmath27 in @xmath146 , both @xmath147 and @xmath58 are contractible ; the intersection of finitely many charts in @xmath146 is either empty or again a chart in @xmath146 .
[ [ proof.-2 ] ] proof .
+ + + + + + for a good triangulation @xmath85 of @xmath148 as above , consider a simplex @xmath86 .
the open star @xmath149 is contained in @xmath60 for some chart @xmath25 in @xmath37 .
let @xmath31 be a point in the interior of @xmath86 , so that @xmath150 .
choose a lifting @xmath151 of @xmath31 in @xmath53 .
then , by lemma [ stars ] and remark [ rems]([contract ] ) , the map @xmath152 is part of a chart for @xmath0 .
the collection of all these charts for open stars @xmath149 is the required atlas @xmath146 .
[ [ section-3 ] ] following the terminology of @xcite , we call a cover @xmath146 of @xmath153 by charts as in proposition [ goodcov ] a _ good cover _ of the orbifold @xmath0 .
let @xmath38 be an orbifold .
recall ( @xcite ) that a sheaf on @xmath0 is given by the following data : for each chart @xmath25 in @xmath37 an ( ordinary ) sheaf of abelian groups @xmath154 on @xmath53 ; for each embedding @xmath155 an isomorphism @xmath156 these isomorphisms are required to be functorial in @xmath43 ; i.e. if @xmath157 is another embedding , then the following square commutes : @xmath158 where ` = ' denotes the canonical isomorphism ; [ action ] it follows that each @xmath154 is a @xmath12-equivariant sheaf on @xmath53 , see @xcite .
[ [ section-4 ] ] with the obvious notion of morphisms between sheaves , these sheaves form an abelian category @xmath159 with enough injectives . referring to the examples in section [ one ] , we remark that if the orbifold is defined from the action of a compact lie group @xmath68 on a manifold @xmath66 , then this category is ( equivalent to ) the category of @xmath68-equivariant sheaves on @xmath66 . and
if the orbifold is defined from a suitable foliation @xmath73 on a manifold @xmath66 , it is ( equivalent to ) the category of holonomy - invariant sheaves on @xmath66 . for a sheaf @xmath7 as above , a _ global section _
@xmath160 of @xmath7 is by definition a system of sections @xmath161 , one for each chart @xmath53 , and compatible in the sense that for each embedding @xmath162 , the identity @xmath163 holds .
the group of all these global sections is denoted @xmath164 .
this defines a functor @xmath165 , into the category of abelian groups , which is right exact and preserves injectives .
for an abelian sheaf @xmath7 , one defines @xmath166 this definition of the sheaf cohomology of @xmath0 is just a special case of the cohomology of a topos @xcite , and hence it satisfies all the standard functoriality and invariance properties . by way of example , we mention some of these properties : any ( strong , @xcite ) map between orbifolds @xmath167 induces an exact functor @xmath168 , and an induced homomorphism @xmath169 .
there is also an ( adjoint ) functor @xmath170 and a corresponding ( ` leray ' ) spectral sequence @xmath171 see ( * ? ? ?
* expos v , section 5 ) .
there are adjoint functors @xmath172 where @xmath173 is the category of abelian sheaves on the underlying space @xmath6 .
there is a corresponding leray spectral sequence @xmath174 where for a point @xmath75 , the stalk @xmath175 is @xmath176 .
( here @xmath177 is the stalk of @xmath154 at any lifting @xmath178 of @xmath31 for some chart @xmath25 , and the stabilizer @xmath179 acts on @xmath177 since @xmath154 is @xmath12-equivariant .
( mayer - vietoris ) if @xmath180 is a union of two open sets , the orbifold structure @xmath0 restricts to orbifold structures @xmath181 and @xmath182 , and there is a mayer - vietoris sequence @xmath183 [ cech - spec ] define a presheaf @xmath184 on the underlying space @xmath6 , by @xmath185 .
for any open cover @xmath146 of @xmath6 , there is a spectral sequence @xmath186 with as @xmath187-term the ech cohomology of this cover @xmath146 with coefficients in this presheaf @xmath188 ; see ( * ? ? ?
* expos v , section 3 ) .
we single out a special case of this last property ( [ cech - spec ] ) .
a sheaf on @xmath0 is said to be _ locally constant _ if each of the ( ordinary ) sheaves @xmath154 is locally constant .
now suppose @xmath146 is a ` good ' atlas for @xmath0 , as in corollary [ goodcov ] , and let @xmath7 be any locally constant sheaf . for any chart @xmath27 in @xmath146 ,
the restriction @xmath189 is a constant sheaf on a contractible space @xmath190 , and @xmath191 is the cohomology of the group @xmath192 with coefficients in the @xmath193-module @xmath194 .
thus , for a locally constant sheaf @xmath7 and a good cover @xmath146 , the spectral sequence takes the form @xmath195
our purpose in this section is to describe explicitly for any orbifold @xmath38 a simplicial set @xmath95 , with the property that any locally constant sheaf @xmath7 on @xmath0 induces a local system of coefficients @xmath5 on @xmath95 for which there is a natural isomorphism @xmath196 see theorem [ main ] below .
let us fix an @xmath36-dimensional orbifold @xmath0 with underlying space @xmath6 and ( chosen ) atlas @xmath37 ; let us also fix a triangulation @xmath85 of @xmath0 , and write @xmath197 for the set of @xmath36-simplices .
we assume that the triangulation is adapted to @xmath37 , as described in section [ trian ] . recall that this means that @xmath85 has the following properties : + [ i ] for each @xmath36-simplex @xmath198 there is a chart @xmath199 , such that @xmath200 ; [ ii ] for each simplex @xmath119 there is a face @xmath201 such that the isotropy is constant on @xmath202 ; in particular every simplex has a vertex @xmath203 with maximal isotropy .
+ we assume that a choice of charts @xmath204 and vertices @xmath203 as in ( [ i ] ) and ( [ ii ] ) above has been made .
( note that we do not require the stronger property of being ` good ' for the triangulation , because in some examples that would force us to construct a simplicial complex which is bigger than necessary , and hence less suitable for calculations . )
we now construct the simplicial set @xmath205 with the same cohomology as @xmath0 .
the description of @xmath95 will use various choices , besides the charts @xmath204 and the vertices @xmath203 already mentioned .
first of all , choose for each simplex @xmath198 a lifting @xmath93 as in lemma [ lift ] , mapped homeomorphically to @xmath200 by @xmath206 .
next , fix for each vertex @xmath207 of @xmath85 a neighborhood @xmath208 of @xmath207 and a chart @xmath209 over @xmath208 , so small that @xmath210 whenever @xmath211 . also fix a lifting @xmath212 , and an embedding @xmath213 with @xmath214 .
we will not require these @xmath208 to belong to the original atlas @xmath37 .
in fact , they can be chosen so small that @xmath215 is also the isotropy group of @xmath207 in the chart @xmath209 , so that the notation is unambiguous ; in this case the lifting @xmath216 is unique .
let @xmath217 be two simplices of maximal dimension @xmath36 , and assume @xmath218 .
below , in section [ five ] , we will construct for any two simplices @xmath119 and @xmath219 with @xmath220 an injective map @xmath221 we will write @xmath222 or @xmath223 if the ( other ) subscripts are clear from the context .
this map @xmath222 in ( [ mu ] ) will not be a homomorphism in general ; it will map @xmath224 to a coset of a conjugate of the subgroup @xmath225 in @xmath226 .
however the construction will have the following multiplicative property : if @xmath227 then , for @xmath228 , @xmath229 moreover , if @xmath230 and @xmath231 then @xmath232 is the identity .
with these choices made , the simplicial set @xmath205 can be described .
as already defined above , @xmath233 furthermore , for @xmath234 , @xmath235 an element of @xmath236 can also be denoted by @xmath237 to suggest the analogy with nerves .
so @xmath238 in ( [ nerve ] ) are @xmath36-simplices , and @xmath239 where @xmath203 is the chosen vertex with maximal isotropy on @xmath240 .
the degeneracy maps @xmath241 @xmath242 are defined in the usual way , @xmath243 the face maps @xmath244 @xmath245 are defined by means of the maps @xmath222 in ( [ mu ] ) , as @xmath246 where the @xmath222 s carry the following subscripts .
write @xmath240 and @xmath247
. then @xmath248 and @xmath249 the latter by ( [ mult ] ) .
the simplicial identities now follow easily .
before we define these maps @xmath222 , we state the theorem .
let @xmath7 be a locally constant sheaf on @xmath0 .
this sheaf induces in a natural way a local system of coefficients @xmath5 on the simplicial set @xmath250 , with , for any @xmath198 , @xmath251 to describe the twisting , observe that , since @xmath7 is locally constant , @xmath252 is constant on @xmath93 .
so for any vertex @xmath253 there is a canonical isomorphism from the stalk at @xmath254 , @xmath255 modulo these isomorphisms , the twisting by an element @xmath256 is now defined as the dashed map in the diagram of maps between stalks @xmath257|>\tip&&&{\cal a}_{{\tilde{u}}_{\sigma_0},\tilde v_0}\ddouble\\ \lambda_1^*({\cal a}_{{\tilde{u}}_{\sigma_1}})_{\raisebox{-1.5pt}{${\scriptstyle { \tilde{v}}}$}}&{\cal a}_{{\tilde{u}},v}\lto\rto^{\overline{g}}&{\cal a}_{{\tilde{u}},v}\rto&\lambda_0^*({\cal a}_{{\tilde{u}}_{\sigma_0}})_{\raisebox{-1.5pt}{${\scriptstyle { \tilde{v}}}$ } } \enddiagram\ ] ] here @xmath258 , so @xmath259 .
furthermore , @xmath260 is the chosen embedding @xmath261 , with the property that @xmath262 .
finally , @xmath263 in the diagram denotes the left action by @xmath264 ( cf .
condition ( [ action ] ) in the description of sheaves in section [ coho ] ) .
[ main ] for any triangulated orbifold @xmath0 with associated simplicial set @xmath95 as above , and for any locally constant sheaf @xmath7 on @xmath0 , there is a natural isomorphism @xmath265 we will now first define the maps @xmath222 involved in the definition of @xmath95 .
the proof of the theorem will be given in section [ proof ] .
fix @xmath266 and @xmath267 as in ( [ subsimp ] ) .
write @xmath268 and @xmath269 for the corresponding vertices with maximal isotropy on @xmath119 and @xmath219 , respectively .
we will construct @xmath270 let @xmath271 be the 1-simplex joining @xmath207 and @xmath272 in @xmath219 .
then the isotropy group of any interior point of @xmath271 agrees with that of @xmath207 , while that of @xmath272 is possibly larger , @xmath272 being maximal on @xmath271 .
recall that we have already chosen @xmath273 first consider the special case that there is a chart @xmath274 for a neighborhood @xmath275 , for which @xmath276 and @xmath277 .
( such a chart exists , for example , when one starts with a ` good ' atlas @xmath37 as in section [ trian ] , in which case one can take @xmath278 . )
choose an embedding @xmath279 with @xmath280 .
let @xmath281 be a lifting of @xmath271 , and choose the other embedding @xmath282 such that @xmath283 . also choose embeddings @xmath284 with @xmath285
. now observe that , since @xmath286 and @xmath287 both map @xmath216 into @xmath288 , one has @xmath289 , and hence there is a @xmath290 such that @xmath291 ( cf .
section [ one ] , remark ( [ two ] ) ) .
similarly , we find @xmath292 and @xmath293 , such that @xmath294 we claim that for any @xmath295 there is a ( unique ) @xmath296 such that @xmath297 to see this , recall first that @xmath83 denotes the group of the chart @xmath298 , and consider @xmath299 note that @xmath300 fixes @xmath301 . since the isotropy along @xmath302 does not decrease from @xmath303 , it follows that @xmath300 fixes @xmath304 , and hence also the point @xmath305 .
thus @xmath306 for some @xmath307 .
now let @xmath308 then @xmath309 , as required for ( [ m ] ) .
we now define @xmath222 in ( [ muvw ] ) by @xmath310 it can be shown that this definition is independent of the choices made above ( of @xmath311 , @xmath312 , @xmath313 ) .
[ [ section-5 ] ] this defines @xmath222 in the special case , where the 1-simplex @xmath271 is contained in a chart @xmath33 with @xmath314 . in the general case , choose a subdivision of @xmath271 into smaller 1-simplices , with vertices @xmath315 choose charts @xmath316 liftings @xmath317 , and embeddings @xmath318 with @xmath319 , and coinciding with the embeddings already chosen for @xmath320 .
furthermore , choose this subdivision of @xmath271 sufficiently fine , and these charts @xmath321 sufficiently small , so that each 1-simplex between @xmath322 and @xmath323 is contained in a chart @xmath324 , as in figure [ chain ] .
@xmath325{\epsfxsize=.8\textwidth\epsffile{chain.ps } } \ ] ] now define @xmath326 exactly as the maps @xmath327 defined in ( [ mukl ] ) above , for @xmath328 , and let @xmath329 be the composition @xmath330 it can be shown that this definition of @xmath222 is again independent of the various choices . in particular , for a finer subdivision of @xmath271 and a refinement of the system of open sets @xmath331 and @xmath332 , one obtains the same map @xmath222 .
we omit the details .
in this section we will apply the construction of the simplicial set @xmath95 from the previous sections to the teardrop orbifold ( as described in @xcite ) , and calculate its cohomology groups . as before @xmath7 denotes a locally constant sheaf of coefficients on the orbifold considered , and @xmath5 denotes the induced local system of coefficients on @xmath95 . the quotient space of the teardrop orbifold is the 2-sphere with one cone point of order @xmath36 . a chart around the cone point consists of an open disk @xmath53 in @xmath333 with structure group @xmath334 ( the finite cyclic group of order @xmath36 ) , which acts on @xmath53 by rotations .
we will denote this orbifold by _ tear - n_. figure [ teardrop ] shows a picture of the quotient space and a triangulation of this orbifold .
the simplices @xmath335 and @xmath336 are on the front and the simplices @xmath337 and @xmath338 are on the back of the teardrop .
moreover , @xmath339 , @xmath340 , @xmath341 and @xmath342 , where @xmath343 .
the atlas we use for this orbifold consists of eight charts : an open disk @xmath344 with a trivial structure group to cover the lower half of the quotient space , and an open disk @xmath53 with structure group @xmath334 ( acting by rotations ) to cover the upper half , and six charts to cover the equator in order to satisfy the compatibility condition for atlases .
note that the triangulation is adapted to this atlas .
figure [ liftings ] shows the liftings ( in the charts ) of the 1- and 2-simplices in the triangulation , as needed for the construction of the simplicial representation for this orbifold , for the case that @xmath345 .
the simplices @xmath346 and @xmath337 are subsets of @xmath60 ; their liftings , denoted by @xmath347 , are shown in the left hand chart @xmath53 .
similarly the liftings of @xmath348 , and @xmath338 are shown in the right - hand chart @xmath344 . using these liftings
we define the maps @xmath351 for every pair @xmath352 and @xmath353 for which the intersection is a 1-simplex @xmath354 and @xmath355 is a vertex with the same isotropy group as the interior of @xmath354 .
note that @xmath356 ( where @xmath352 , @xmath353 and @xmath357 are three different simplices ) is trivial except for @xmath358 .
so the only possibly non - trivial @xmath222-maps are those with codomain @xmath359 . in order to construct @xmath360 note that @xmath271 , as used in section [ five ] , is @xmath361 .
we choose @xmath362 and @xmath363 , so both @xmath364 and @xmath365 , as used in section [ five ] , are the identity map .
moreover , we choose @xmath366 ( notation as in section [ simpset ] ) , whereas @xmath367 and @xmath368 have to be the same , because the liftings of @xmath369 in @xmath370 and @xmath371 are the same point in @xmath53 . and
it is obvious that @xmath372 also ( by the multiplicative property ( [ mult ] ) of @xmath222 ) , @xmath373 for similar reasons , when we choose @xmath374 we find that both maps @xmath375 are described by @xmath376 the only non - trivial @xmath222-maps are @xmath377 , and @xmath378 from @xmath379 into @xmath380 . with the notation of section [ simpcom ] , @xmath381 and
we choose @xmath382 with @xmath383 as in figure [ liftings ] .
moreover , we take @xmath384 then ( [ een ] ) and ( [ twee ] ) above induce that @xmath385 it follows from the choice of the liftings in figure [ liftings ] , that @xmath386 where @xmath219 is rotation over @xmath387 , which generates @xmath388 . we conclude that the group elements @xmath389 and @xmath390 are all the identity element , and @xmath391 , so : @xmath392 similarly one can show that : @xmath393 ( this also follows from the fact that they have to be each other s inverses by the multiplicative property of @xmath222 as stated in ( [ mult ] ) . ) with these @xmath222-maps the simplicial complex can be described as follows . @xmath394
the degeneracy maps @xmath395 are trivial as described in section [ simpset ] , and the face maps are straightforward compositions on all combinations which do not contain neighboring @xmath396 and @xmath397 in their @xmath86-part , since the @xmath222-maps are trivial in these cases . for the @xmath398-combinations
, we have to use the non - trivial part of the @xmath222-maps described above .
for example , deleting the sigma s from the notation , we can describe the face operators on @xmath399 by : @xmath400 so for @xmath401 , we find that @xmath402 .
we use this in the following calculation , @xmath403 ( this calculation illustrates the effect of the fact that a loop around the singular point @xmath404 is contractible via the lower half of the teardrop . ) to find the second cohomology group , we have to do some cocycle - coboundary calculations in the local system of coefficients @xmath5 on the simplicial set @xmath95 . a cohomology class in @xmath406 is represented by a cocycle @xmath407 , which we write as an @xmath5-valued map @xmath408 .
since the second cohomology group of the plane is zero , we can choose the representant @xmath312 in such a way that @xmath409 when @xmath410 and moreover , + | for any orbifold @xmath0 , we explicitly construct a simplicial complex @xmath1 from a given triangulation of the ` coarse ' underlying space together with the local isotropy groups of @xmath0 .
we prove that , for any local system on @xmath0 , this complex @xmath1 has the same cohomology as @xmath0 .
the use of @xmath1 in explicit calculations is illustrated in the example of the ` teardrop ' orbifold . |
the discovery of high - temperature superconductivity in the cuprate materials has sparked huge interest in models of strongly correlated electrons in low - dimensional systems , such as the hubbard , @xmath2-@xmath3 and @xmath2-@xmath0 models .
these models are exactly solvable in one dimension , at least in some special cases ; but the two - dimensional models pose a formidable numerical challenge .
the ` minus sign ' problem is a major stumbling block for monte carlo calculations in these fermionic systems ; and the convergence of density matrix renormalization group ( dmrg ) calculations is slow in two dimensions .
exact finite - lattice calculations are limited to small lattice sizes ; while series expansions have typically only been useful for special cases such as the half - filled limit . in these circumstances ,
a considerable effort has been invested in the study of ` ladder ' systems consisting of two or more coupled chains of sites .
ladders provide a ` half - way house ' , in some sense , between one and two dimensions ; and they also display some very interesting effects in their own right @xcite .
they display quite different behaviour depending on whether the number of legs is even or odd , as in the haldane effect for the heisenberg ladders @xcite .
furthermore , experimental compounds have been found which form ladders @xcite , such as srcu@xmath4o@xmath5 @xcite , which may allow the theoretical predictions to be tested experimentally .
the @xmath2-@xmath3 model is an ` effective hamiltonian ' for the parent hubbard model , valid when the coulomb repulsion is large , but nowadays it is considered as an interesting model in its own right @xcite .
the @xmath2-@xmath0 model is a variant in which the rotational symmetry is broken , and the spin interactions are ising - like .
the two - leg @xmath2-@xmath3 ladder has been extensively studied , using exact diagonalization @xcite , quantum monte carlo @xcite , the dmrg technique @xcite , a combination of different methods @xcite , or using mean - field or approximate analytic methods @xcite . near half - filling
, the model has been explored using dimer series expansions @xcite .
our object is to study the corresponding two - leg @xmath2-@xmath0 ladder , and compare the results for the two models .
this model has not been studied before , as far as we are aware , but the @xmath2-@xmath0 chain has been discussed by batista and ortiz @xcite , and the @xmath2-@xmath0 model on the square lattice has been treated by several groups @xcite .
our primary tool is the dmrg approach , supplemented with a few series calculations near half - filling .
the phase diagram for the @xmath2-@xmath3 ladder has been discussed by troyer et al .
@xcite , hayward and poilblanc @xcite , and m " uller and rice @xcite . at large @xmath6
, the holes all clump together , and phase separation occurs into hole - rich and hole - poor regions . at intermediate @xmath6 , near half - filling , a ` c1s0 ' or luther - emery phase occurs , where the spin excitations are gapped , while there is a gapless charge excitation mode .
troyer et al .
@xcite found evidence of pairing between the holes in this region , together with long - range superconducting correlations with modified d - wave symmetry .
the spin gap is discontinuous at half - filling , as the simple magnon excitation gives way to a particle - hole excitation with spin . at smaller @xmath6
, the phase structure appears to become more complicated , with a possible c2s2 phase appearing ( two gapless charge modes and two gapless spin modes ) @xcite ; while at extremely small @xmath6 , a nagaoka phase is expected to appear @xcite , where each hole is surrounded by a region of ferromagnetic spins , forming a ferromagnetic polaron . in that region
no spin gap occurs , and the holes repel each other .
the @xmath2-@xmath0 ladder might be expected to show similar behaviour .
the major difference between the models is that quantum spin fluctuations are absent in the @xmath2-@xmath0 model , and the system exhibits long - range antiferromagnetic order for half - filling at @xmath7 , whereas the @xmath2-@xmath3 model does not .
this long - range order will be destroyed at any finite temperature , however , and both models will then display similar long - range antiferromagnetic correlations .
the two models should be very similar in most other aspects .
this expectation is borne out by our numerical results .
the phase diagram for the t-@xmath0 ladder looks very similar to that of the t - j ladder , except that the critical couplings are about twice as large , and the tomonaga - luttinger c1s1 phase extends to somewhat higher electron densities . in section [ secii ]
we specify the model , and consider its behaviour in various limiting cases . in section [ seciii ] a brief discussion of the dmrg method is given , and in section [ seciv ] our numerical results are presented .
our conclusions are given in section [ secv ] .
the hamiltonian of the @xmath2-@xmath0 ladder model is ( see figure [ fig1 ] ) @xmath8 -@xmath3 ladder . ] here the index @xmath9 labels the two legs of the ladder , @xmath10 labels the rungs of the ladder , the couplings @xmath11 are the strengths of the spin interactions on legs and rungs respectively , and @xmath12 are the hopping strengths on legs and rungs .
the projection operators @xmath13 forbid double occupancy of sites as usual . a density - density interaction term is sometimes included as a relic of the parent hubbard model , but we do not do that here . in the half - filled case , with a single electron occupying every site ( n=1 ) , the model becomes equivalent to a simple classical ising antiferromagnet .
the ground state is a doubly degenerate antiferromagnetic state ( fig .
[ fig2]a ) , with energy @xmath14 excitation ; c ) a domain wall ( ` soliton ' ) excitation . ] where @xmath15 is the number of rungs of the ladder .
the system can be solved exactly in various limiting cases : for @xmath17 , the system consists of independent dimers on the rungs of the ladder .
the eigenstates on a single rung are listed in table [ table1 ] .
the ground state is doubly degenerate here , with both ` singlet ' and ` triplet ' states with @xmath18 having the same energy , unlike the case of the full @xmath2-@xmath3 ladder .
this degeneracy means that one can not simply compute a perturbation series expansion about the dimer limit in this case , unless one employs degenerate perturbation theory , or introduces an ` artificial ' interaction to lift the degeneracy .
.rung dimer eigenstates .
[ cols="^,^,^,^,^",options="header " , ] [ table1 ] in this case we end up with two independent chains .
the chain behaviour has been discussed by batista and ortiz @xcite .
they showed that the spins were n ' eel ordered along the chain in the ground state , which could then be mapped onto an xxz spin chain .
phase separation occurs at the ( rather large ) value @xmath20 for all hole densities . below that value ,
the system forms a gapless luttinger liquid , with correlation exponent @xmath21 and charge velocity @xmath1 .
the system can be exactly solved via the bethe ansatz , at quarter filling giving @xmath22 where @xmath23 .
thus for @xmath24 , we have @xmath25 , while for @xmath26 , we have @xmath27 , implying that superconducting correlations dominate at large distances . at half filling , on the other hand , we have free fermion ( metallic ) behaviour with @xmath28 . in this limit ,
the model becomes equivalent to a classical , static ising model .
unless otherwise stated , we assume periodic boundary conditions , and an even number of rungs l. the ground state at half - filling is the fully antiferromagnetic state shown in figure [ fig2]a , with @xmath30 . the low - energy spin excitations will consist either of localized flipped spins ( ` magnons ' ) with @xmath31 , as shown in fig .
[ fig2]b , or else domain wall ( ` soliton ' ) excitations , as shown in fig .
[ fig2]c , analogous to the ` spinon ' excitations in heisenberg chains , but carrying @xmath30 in this case .
the total number of solitons , in the absence of holes , and assuming periodic boundary conditions , has to be even for l even , and odd for l odd , as in the heisenberg chain .
the energy gaps for these two excitations are respectively @xmath32 since there are no quantum spin fluctuation terms in the hamiltonian , these spin excitations are of course static , in the absence of holes .
note that since the solitons carry integer spin , there is no possibility of spin - charge separation in this model .
the lowest - energy charge excitation will consist of a single hole in the antiferromagnetic background ( fig .
[ fig3]a ) , with spin @xmath33 , and energy @xmath34 the lowest eigenstates in the 2-hole sector consist of a bound pair on adjacent sites ( figs .
[ fig3]b,[fig3]c ) , with spin @xmath35 and energies @xmath36 it is clear that holes will cluster together to minimize the number of ` broken bonds ' and hence the energy , and the system will be phase separated in this ising limit . for small but finite @xmath37 , one can study the system via perturbation series calculations in @xmath37 .
some results of these calculations will be shown in later sections .
the single hole states illustrated in fig .
[ fig3]a are localized states , because there is an energy barrier preventing them from hopping : any hop will disturb the antiferromagnetic alignment of the spins . at higher energy ,
however , there will be states such as that shown in figure [ fig4 ] , with unperturbed energy @xmath38 where the hole is free to hop along the zig - zag path shown without incurring any further penalty in spin interaction energy .
assuming periodic boundary conditions , these states are made up of a hole plus a soliton for l even , or a hole alone for l odd . for finite @xmath37 , they will form a band of itinerant electron states .
the mobility of the electrons in both the chain and the ladder systems is a major point of distinction from the two - dimensional model , where the holes are ` confined ' , i.e. can not move without creating a ` string ' of overturned spins behind them , and paying a penalty in spin interaction energy @xcite .
the states ( 3a ) and ( 4 ) will mix as soon as @xmath2 is turned on , and both configurations should be regarded as hole - soliton bound states .
we will also employ open boundary conditions in our calculations . in this case
, the holes will cluster towards the boundaries in the ising limit , in order to minimize the number of broken bonds .
in this case the ground - state energy is @xmath39 and the lowest 1-hole state ( fig .
[ fig4a]a ) has energy @xmath40 while the lowest 2-hole state ( fig .
[ fig4a]b ) has energy @xmath41 the single hole state is not necessarily accompanied by a soliton in this case . in this limit
, a single hole will move through the lattice as a free particle , and its dispersion relation is naively expected to be @xmath43 for simplicity , we shall restrict our remarks henceforth to the isotropic case @xmath44 , when the 1-hole energy gap in this limit is @xmath45 .
the behaviour at small @xmath6 should be similar to that of the @xmath2-@xmath3 model .
the 1-hole energy in the two - dimensional @xmath2-@xmath3 model has been discussed by a number of authors @xcite . at extremely small @xmath6 ( of order @xmath46 , in the region of the nagaoka phase @xcite ) , the hole will be surrounded by a ferromagnetic ` polaron ' , or region of ferromagnetically aligned spins , with spin energy of order @xmath47 for a disc of radius @xmath48 , within which the hole is fully mobile . in this regime , the 1-hole energy is @xmath49 where @xmath50 is a numerical constant . at larger @xmath6 , however , the lowest - energy states correspond to brinkman - rice ` string ' states @xcite , where the hole is confined by a ` string ' of overturned spins within an antiferromagnetic background .
this picture gives a 1-hole energy @xmath51 where @xmath52 .
shraiman and siggia @xcite estimated @xmath53 and @xmath54 for the @xmath2-@xmath0 model , quite close to the values found numerically by barnes et al @xcite , and white and affleck @xcite . for a one - dimensional chain ,
no nagaoka phase occurs , but for a ladder the condition is once again met that holes can hop around closed loops , and a nagaoka phase is expected @xcite .
white and affleck @xcite have pointed out that the ferromagnetically aligned spins inside the polaron will orient themselves in the xy plane , rather than along the @xmath55 axis , because this costs an exchange energy of @xmath56 per bond , rather than @xmath57 .
for our ladder model , the ferromagnetic ` polaron ' region will be one - dimensional .
if the polaron region spans l rungs of the ladder , the cost in magnetic energy will be @xmath58 assuming the 1-hole wavefunction to vanish at the edges of the polaron , it will take the form @xmath59 where @xmath60 denotes a hole at rung @xmath61 on leg @xmath62 of the ladder , and @xmath63 is a normalization constant .
the kinetic or hopping energy is then easily found to be @xmath64 minimizing the total energy @xmath65 with respect to @xmath15 we find @xmath66 for intermediate @xmath6 , the brinkman - rice string picture @xcite will give a qualitatively similar behaviour @xmath67 where @xmath68 .
it is doubtful whether approximations such as those used by shraiman and siggia @xcite to calculate the coefficients @xmath69 and @xmath70 for the 2d model are at all accurate for the ladder , and we will not attempt an explicit calculation of these coefficients . in the regime of physical interest ,
the @xmath2-@xmath3 ladder is believed to be in a c1s0 or luther - emery phase , with gapped spin excitations and gapless charge excitations corresponding to bound hole pairs .
several authors @xcite have discussed an effective hamiltonian to describe these bosonic excitations , which would capture the low - energy physics of the model .
here we merely summarize their results . in the recent analysis of white , affleck and scalapino @xcite , they use a bosonization technique to construct the low - energy effective hamiltonian @xmath71 \label{eq11a}\ ] ] where @xmath72 is a chemical potential , @xmath73 is the momentum density conjugate to @xmath74 , @xmath1 is the velocity of the corresponding gapless low - energy excitations and the parameter @xmath21 controls the correlation exponents .
the two parameters @xmath1 and @xmath21 must be extracted from numerical data .
white et al .
@xcite show that the finite - size scaling behaviour of a general low - energy excitation is @xmath75 \label{eq11b}\ ] ] where @xmath76 is the ground - state energy for a given density @xmath61 , @xmath77 and @xmath78 are occupation numbers for left and right moving states of momentum @xmath79 , @xmath15 is the number of rungs of the ladder , and @xmath80 and @xmath81 are integer - valued quantum numbers .
the total charge relative to the ground state is @xmath82 , and the other quantum number @xmath81 measures the chiral charge " . by measuring the ground - state energies for three different charge states with @xmath83
, one can determine the ratio @xmath84 : @xmath85 this is directly related to the electron compressibility @xmath86 of the two - leg ladder , generally defined as @xmath87 using ( [ eq11c ] ) , where @xmath61 is the electron density .
this formula applies for either periodic or open boundary conditions , in principle .
the velocity may be measured independently using the excitation energy of the lowest state of momentum @xmath88 for periodic bc @xmath89 for open bc , the momentum spacing is @xmath90 , and the corresponding formula is @xmath91 hence one can obtain estimates for @xmath21 and @xmath1 separately .
one may also use ` twisted ' boundary conditions with the wave function acquiring a phase @xmath92 at the boundary , corresponding to an aharonov - bohm flux threading the one - dimensional ring formed by joining the ends of the ladder together .
this increases the ground - state energy by @xmath93 and allows one to determine the combination @xmath94 .
this approach was used by hayward and poilblanc @xcite . finally , white et al .
@xcite discuss how one may estimate @xmath21 from the decay of friedel oscillations in the system .
friedel oscillations are density oscillations produced near the boundary of an open ladder , which decay with a power law corresponding to the exponent @xmath21 with distance into the ladder . using their bosonization analysis and a conformal transformation ,
white et al @xcite show that for a system of length @xmath15 the density at site ( rung ) @xmath95 should vary as @xmath96^{k_{\rho } } } \label{eq11h}\ ] ] where @xmath97 are constants , and @xmath98 is the minimum wave vector of friedel oscillations in the c1s0 phase , corresponding to two holes per wavelength .
this is the same wave vector that would occur for a one - component spinless hard core bose gas , made up of tightly bound hole pairs .
schulz @xcite has argued in the case of the @xmath2-@xmath3 ladder that @xmath21 should take the universal value @xmath99 at half - filling ; but his argument starting from the rung dimer limit is not necessarily applicable to the @xmath2-@xmath0 model .
note that the picture may be complicated by the occurrence of a gapped charge density wave ( cdw ) phase at special commensurate filling factors such as @xmath100 or @xmath101 . in this case , the charge density oscillations persist at all distances from the walls , corresponding to broken translational symmetry and a regular pattern of hole placements .
this possibility has been discussed by riera et al @xcite and white et al @xcite .
the model ( [ tjzham ] ) is solved using the density matrix renormalisation group ( dmrg ) method @xcite .
calculations have been performed at both half - filling and quarter - filling .
calculations have been performed with both periodic and open boundary conditions in the horizontal direction , but most of the results use obcs because the convergence is much better in this case . in the half - filled case
we have calculated the ground state as well as one , two and four - hole excited states . in the one - hole case
we have calculated @xmath102 , 3/2 and 5/2 states .
we have calculated energies of these states as well as density profiles @xmath103 .
the infinite lattice algorithm @xcite is employed in this case , using typically @xmath104 states per block and symmetry sector , which corresponds to a total of about 300 system block states .
the lattice build - up phase is illustrated in fig.[infinite_lattice ] . by running tests for a number of different values of @xmath105
, we established that the energies of these states have been resolved with sufficiently high accuracy for our purposes . in the quarter - filled case , it is necessary to use the finite lattice method @xcite in order to achieve reasonably converged energies .
we use the infinite lattice method to build a ladder to a given size , increasing the system block size by a whole rung at each dmrg step as in fig .
[ infinite_lattice ] .
we then perform a finite lattice sweep at the fixed lattice size , using the previously developed system blocks to improve the accuracy of the calculation .
once a target size has been completed with a finite lattice sweep , the infinite lattice algorithm is resumed to proceed to a larger target lattice size whereupon the sweeping procedure is again employed . this way we obtain accurate finite lattice method results for a number of lattice sizes , e.g. , @xmath106 , 8 , 12 , 16 , 24 , 32 , 64 , 128 , 256 lattice sites .
wherever necessary , the data are extrapolated to the bulk limit @xmath107 by fitting a low - order polynomial in @xmath108 to the finite - lattice data .
we have calculated numerical dmrg results for this model for the isotropic case , @xmath109 , and for various @xmath110 , on lattices of l rungs with l even .
our discussion henceforth will be limited to this case .
we begin by considering states with a finite number of holes doped into the half - filled system ( n=1 ) .
figure [ fig6 ] shows dmrg estimates of the energy gap for a single hole with spin @xmath102 , as a function of @xmath110 , for both periodic and open boundary conditions .
the results of a series calculation are also shown for the periodic case .
the series estimates were obtained assuming the nagaoka form ( [ eq10c ] ) @xmath111 accordingly , the series in @xmath2 for @xmath112 was euler transformed @xmath113 to bring the singularity to @xmath114 ; then a further change of variable was made @xmath115 so that one may treat the function @xmath116 as analytic in @xmath117 near @xmath118 ; and finally , differential approximants were used to extrapolate this function to @xmath118 .
it can be seen that the series results agree very well with the dmrg results for @xmath119 , but run a little lower below that . for one hole , as a function of @xmath110 .
filled circles , open boundary conditions ; open squares , periodic boundary conditions ; dashed line , estimates from ising series expansion . ]
it can also be seen that there is a difference between the dmrg results with periodic and open boundary conditions . as discussed in section [ secii.3 ] , this is because the state with periodic boundary conditions is actually a hole - soliton bound state .
the difference between the two energies is shown in the inset to figure [ fig6 ] .
-@xmath3 ( left ) and @xmath2-@xmath0 ( right ) model on a 4-rung ladder with 7 electrons . ] at smaller @xmath110 , one is tempted to look for a transition to a nagaoka phase @xcite , i.e. a phase with ferromagnetic spin order or at least a ferromagnetic polaron bubble around the hole . in the @xmath2-@xmath0 model
, however , the situation is rather different from the @xmath2-@xmath3 model , where the full spin - rotation symmetry applies .
the scenario is illustrated in figure [ fig8 ] , where we show the complete 1-hole energy spectrum of the @xmath2-@xmath3 and @xmath2-@xmath0 models on a 4-rung cluster .
the left - hand panel shows the @xmath2-@xmath3 case , where at @xmath120 the ground state has maximal spin @xmath121 ( ferromagnetic ) .
as @xmath3 increases , at some critical coupling @xmath122 this state crosses another state with minimal spin @xmath123 , which becomes the ground state thereafter .
the right - hand panel shows the @xmath2-@xmath0 case . at @xmath124 ,
the system is rotation symmetric , the nagaoka theorem applies and the ground state again has the maximal possible spin @xmath125 and degeneracy @xmath126 .
as soon as @xmath0 becomes finite , however , the symmetry is broken , the multiplet splits into its @xmath127 components , and the state with minimal @xmath128 becomes the ground state . clearly , there is no level crossover in the @xmath2-@xmath0 case .
all we can expect for increasing @xmath0 is a continuous fading - out of the ferromagnetic correlations .
this accords with the smooth approach of the 1-hole energy to @xmath129 as @xmath130 in figure [ fig6 ] .
a comparison of the numerical data at small @xmath110 with the theoretical prediction ( [ eq10c ] ) gives the following results .
a dlog pade analysis of the series gives a rather inaccurate estimate of the exponent in the range 0.5 - 0.7 , which would be consistent with 2/3 .
a fit to the dmrg data over the range @xmath131 , however , gives @xmath132 with an exponent much closer to @xmath133 than @xmath134 .
this is illustrated by the difference between the dmrg data and the series extrapolation in figure [ fig6 ] at small @xmath110 .
thus the data do not seem to behave in accordance with theoretical expectations in this instance .
figure [ fig10 ] shows the bandwidth for the 1-hole state with @xmath135 predicted by the ising series expansion , as a function of @xmath110 .
we see that the predicted bandwidth rises as @xmath110 decreases , as one might expect , until at @xmath136 it reaches a peak , and begins to drop towards zero .
similar behaviour in the region of small @xmath6 has been predicted for the 2d @xmath2-@xmath3 model by martinez and horsch @xcite and liu and manousakis @xcite , and confirmed by series calculations @xcite .
the theoretical predictions assume long - range antiferromagnetic order , and construct an effective hamiltonian in which the spin dynamics are treated in linear spin - wave theory , and the holes are treated as spinless fermions , following schmitt - rink , varma and ruckenstein @xcite .
the results are consistent with the string picture of brinkman and rice @xcite , and appear to show a sequence of narrow ` string ' excitation peaks in the spectral function , corresponding to holes bound in a linear potential . the effects are even more marked in the @xmath2-@xmath0 model . as a function of @xmath110 , estimated from the ising series expansion . ]
it would be very interesting to check the actual behaviour of the bandwidth by means of dmrg estimates , but unfortunately our codes can not distinguish different momentum eigenstates . in studying 2- and 4-hole states and looking for the phase separation boundary
, we wanted to avoid the holes sticking to the boundary and thus obscuring the bulk physics .
accordingly , we have added an extra ` boundary potential ' term at the sites adjacent to the boundary : @xmath137 as it turns out , this makes a negligible difference to the energy gaps in the region of interest . in the 2-hole sector , the first quantity of interest is the 2-hole binding energy @xmath138 figure [ fig11 ] shows the binding energy as a function of @xmath110 for both @xmath18 and @xmath139 , computed with open boundary conditions .
it can be seen that a 2-hole bound state with @xmath18 forms for large @xmath110 , but the binding energy vanishes at @xmath140 , and below this point the holes do not bind . in the @xmath141 channel
, there is a small binding energy at large @xmath110 , which vanishes at @xmath142 .
, calculated with obc and finite boundary potentials . filled circles : @xmath18 : filled squares : @xmath139 .
right : average electron number per rung as a function of distance along the ladder , for 2-hole states on a lattice of 34 rungs : a ) @xmath143 ; b ) @xmath144
. the lines are fits using equation ( [ eq12a ] ) . ]
similar results have been found for other members of this family of models .
for the 2d @xmath2-@xmath0 model , riera and dagotto @xcite estimated that the holes become unbound at @xmath145 ; while for the @xmath2-@xmath3 ladder , jurecka and brenig @xcite suggest that hole binding vanishes at @xmath146 .
there is binding in the @xmath147 channel for the @xmath2-@xmath3 ladder , but it is considerably smaller than in the @xmath148 channel . wherever the 2-hole binding energy is positive ,
we expect a continuous band of 2-hole bound states to appear above the ground state ; so that within the 2-hole sector , particle - hole excitations ( ` excitons ' ) can occur down to zero energy .
in other words , excitons with @xmath18 are gapless , as found by troyer et al @xcite for the @xmath2-@xmath3 ladder .
we note that the spin energy gap between the @xmath139 and @xmath18 2-hole states also drops to zero at @xmath149 .
it always remains below the @xmath139 ` magnon ' energy @xmath150 ; or in other words , an exciton with @xmath151 in the 2-hole sector has an energy substantially below that of the magnon .
therefore the spin gap drops discontinuously as one moves away from half - filling , which again agrees with the behaviour found by troyer et al .
@xcite for the @xmath2-@xmath3 ladder .
the right panels of figure [ fig11 ] show profiles of the electron density in the 2-hole state as a function of distance along the ladder , calculated for a lattice of 34 rungs with open boundary conditions . at low @xmath110 ,
the two holes are unbound and separated from each other , forming two distinct troughs in the density profile ( fig .
[ fig11 ] ) . at @xmath152
a sharp crossover takes place , where the two holes bind together and form a single trough in the density profile .
as it happens , both profiles can be fitted quite well with a friedel oscillation form @xmath153 } { [ ( l+1)\sin[\pi j /(l+1)]]^{k_{\rho } } } \label{eq12a}\ ] ] where @xmath154 and @xmath155 in case a ) and @xmath156 and @xmath157 in case b ) . ,
calculated with obc and finite boundary potentials .
insets : electron density profiles at characteristic values of @xmath110 , together with fits to the corresponding friedel expression . ]
the 4-hole binding energy for @xmath18 is shown as a function of @xmath110 in figure [ fig13 ] .
it is finite at large @xmath110 , and vanishes smoothly at @xmath158 .
this is taken to mark the point where phase separation occurs .
the smooth behaviour of the binding energy may indicate a kosterlitz - thouless transition at this point .
the insets show examples of the electron density profiles as functions of distance along the ladder for the 4-hole ground state . for @xmath159 , we see four separated holes ; for @xmath144 , we see two separated hole - pairs ; and for @xmath160 , we see a single 4-hole cluster , reinforcing the picture given above .
note that the friedel form ( [ eq12a ] ) again fits the density profiles remarkably well , upon choosing the appropriate value for @xmath61 .
the fits in the pair binding region indicate that the parameter @xmath21 lies approximately in the range 0.8 - 1.0 near half - filling . , at @xmath161 , @xmath162 and @xmath163 . ]
next , we attempt to measure the isothermal compressibility in the non - phase - separated region , whose inverse is @xmath164/4 \label{eq13}\ ] ] where @xmath165 is the number of sites , and @xmath166 the electron density .
figure [ fig14 ] shows @xmath167 as a function of @xmath108 , for some couplings @xmath168 , fitted by a quadratic polynomial in @xmath108 .
the extrapolations to the bulk limit @xmath107 are compatible with zero , indicating that the inverse compressibility vanishes at half - filling .
this seems surprising at first sight , but the same result applies for the 1d @xmath2-@xmath3 chain at the integrable point @xcite .
siller et al .
@xcite also found a tiny result for the inverse compressibility of the isotropic @xmath2-@xmath3 ladder near half - filling .
the result can be understood as follows .
the dispersion relation in the conduction band flattens out near the band edge , and hence the dispersion relation of a single hole at half filling is quadratic in momentum @xmath169 , as illustrated by the free fermion equation [ eq7 ] or the calculation of sushkov @xcite for the t - j ladder .
the dispersion relation for a bound hole - pair bosonic excitation will likewise be quadratic in @xmath169 .
the vanishing of any term linear in @xmath169 implies that @xmath170 and eq .
then leads to @xmath171 .
next we consider states at or near the quarter - filling point ( @xmath172 ) .
reliable results are very hard to obtain for @xmath173 .
phase separation presumably occurs in this region , so that the ground state becomes inhomogeneous , and very difficult to treat by dmrg methods . for the most part
, our results will be for smaller @xmath110 .
/2 $ ] at quarter - filling , as a function of @xmath110 . ]
figure [ fig15 ] shows the energy gap between states with odd and even numbers of holes , @xmath174/2 $ ] , where @xmath175 is the number of holes , as a function of @xmath110 .
it can be seen that for @xmath110 less than about @xmath176 , this gap is small and positive , but is consistent with zero beyond that point , within substantial errors .
this behaviour is very similar to the two - hole gap shown in figure [ fig18 ] .
in fact it appears that the energy surface is quite smooth in this region , with little if any ` corrugation ' between states with even or odd numbers of holes .
this indicates that pairing effects between holes are weak or nonexistent in this region .
$ ] at quarter - filling , as a function of @xmath110 . ]
figure [ fig16 ] shows the spin gap @xmath177 $ ] as a function of @xmath110 .
the insert shows examples of the finite - size scaling behaviour of this quantity as a function of @xmath108 , and the linear extrapolations made to the bulk limit .
it can be seen that the results are consistent with zero over the whole region @xmath178 .
convergence for @xmath179 is not very good .
thus for @xmath180 , we appear to be in the c1s1 phase discussed by hayward and poilblanc @xcite for the t - j ladder , where both the charge gap and the spin gap vanish .
/2 $ ] at quarter - filling , as a function of @xmath181 , at selected values of @xmath110 . ]
/2 $ ] at quarter - filling , as a function of @xmath110 . ]
figure [ fig17 ] shows the two - hole energy gap @xmath182 $ ] as a function of @xmath108 at selected values of @xmath110 . for @xmath183 ,
the gap appears to scale to a finite value in the bulk limit @xmath184 , whereas the limit is consistent with zero from there on .
figure [ fig18 ] shows the estimated bulk limit as a function of @xmath110 .
the finite charge gap for @xmath185 corresponds to a cusp in the energy density versus filling factor surface , and is evidence of the formation of a cdw state at this commensurate filling factor , according to white _ et al .
_ @xcite . ;
squares : @xmath186 . case a ) : @xmath187 ; b ) : @xmath188 ; c ) : @xmath189 ] we have looked for further evidence of the cdw state in the electron density profiles and correlation functions . the density profile on an odd - rung lattice at quarter filling shows some evidence of rung - to - rung oscillations , but they die away as @xmath190 , and so do not seem to correspond to the cdw .
examples of the density - density correlation functions are shown in figure [ fig19 ] , with one site fixed at the middle of the ladder , and the distance to the other site varying , including both sites on the same leg and those on the opposite leg of the ladder .
there is a positive correlation for nearby sites , and a friedel rung - to - rung oscillation near the boundary . in the intermediate region
, there is no strong evidence of either a rung - to - rung alternation or a checkerboard alternation .
one major qualitative difference is obvious , however , between figures [ fig19]a ) and [ fig19]c ) . at @xmath191 ,
the correlation function is virtually zero at intermediate and large distances , as one would expect of a solid phase : this is consistent with the cdw behaviour . for @xmath192 ,
on the other hand , the correlation function has a substantial dip at intermediate distances , which actually increases with increasing lattice size .
this behaviour is more reminiscent of a liquid or a gas .
ladder . ]
as expected , the phase behaviour of the t-@xmath0 ladder at @xmath7 appears to be broadly similar to its counterpart , the t - j ladder .
a schematic diagram is shown in figure [ fig21 ] . at half - filling ,
the system is equivalent to an ising antiferromagnet , with long range magnetic order ; but at any finite temperature or hole density , the order parameter will vanish , leaving only finite - range antiferromagnetic correlations . near half - filling , phase separation into hole - rich and hole - poor regions occurs beyond @xmath193 , which is about half the critical value for the t-@xmath0 chain @xcite , but twice the critical value for the full t - j ladder @xcite . below that , there is a c1s0 or luther - emery phase , where the holes bind into pairs and the charge gap vanishes , but the spin gap remains finite .
the spin gap is discontinuous at half - filling , because the exciton spin gap is less than the magnon gap .
these features are identical with those found for the t - j model @xcite .
we have not explored the question whether there are superconducting correlations in this region , but they are expected to be present if @xmath194 . the binding between hole pairs vanishes in the @xmath195 channel below @xmath142 , and in the @xmath30 channel below @xmath152 .
below that , the holes repel each other , and we presumably enter the c2s2 gapless phase discussed by m " uller and rice @xcite for the t - j case ( although exact diagonalization data up tp @xmath196 are somewhat equivocal as to whether the @xmath197 states are degenerate ) .
these different regimes are clearly evident in the hole density profiles on a finite lattice . in the nagaoka limit as @xmath198
, a finite - lattice calculation reveals rather different behaviour in the two models in the one - hole sector . in the t - j case , a state with maximal spin @xmath199 crosses over the @xmath123 state to become the ground state as @xmath200 . for the t-@xmath0 system ,
on the other hand , the @xmath201 state remains the ground state at all @xmath110 , and the nagaoka limit is approached smoothly .
states with higher @xmath127 become degenerate with the @xmath201 state in the limit , to form a multiplet with maximum total spin @xmath199 .
any ferromagnetic correlations in the ground state appear to be transverse to the @xmath55 direction , as argued by white and affleck @xcite .
our dmrg results for the one - hole gap appear to decrease as @xmath202 , rather than @xmath203 as predicted by theory ; the theory needs re - examination for this case .
some crude estimates of the effective hamiltonian parameters in the luther - emery phase were made at half - filling .
friedel - type fits to the density profiles give @xmath204 , although it is not clear that the schulz argument @xcite that @xmath205 at half - filling is applicable to the t-@xmath0 case .
we found that @xmath206 in that limit , a general result which should apply to chains and ladders for both models .
we have also made some studies of the quarter - filled case .
for @xmath207 , the system appears to be in a cdw state , with a finite two - hole gap and solid - like density - density correlation functions , although we could not discover any particular pattern of charge density waves .
this behaviour was predicted for a commensurate filling factor by white , affleck and scalapino @xcite .
we find that the spin gap vanishes in this phase , and in fact is compatible with zero at all couplings . for @xmath180 ,
any pairing effects are weak or nonexistent , and the system appears to be in a c1s1 tomonaga - luttinger phase with a single gapless spin and single gapless charge mode , similar to that discussed for the t - j ladder at densities at or below quarter - filling by hayward and poilblanc @xcite .
the calculations become very unstable beyond @xmath208 , and we take this as indicating the onset of phase separation again at larger @xmath110 . at very small electron densities , an electron - paired phase should exist . these findings
are indicated schematically in figure [ fig21 ]
. this work forms part of a research project supported by a grant from the australian research council .
we are grateful for extensive computational support provided by the australian partnership for advanced computing ( apac ) national facility and by the australian centre for advanced computing and communications ( ac3 ) .
rice , z. fr physik b103 , 165 ( 1997 ) .
e. dagotto and t.m .
rice , ` science ' 271 , 618 ( 1996 ) .
haldane , phys .
50 , 1153 ( 1983 ) . z. hiroi and m. takano , nature ( london ) * 377 * , 411 ( 1995 ) .
m. azuma , z. hiroi , m. takano , k. ishida and y. kitaoka , phys .
rev . letts .
73 , 3463 ( 1994 ) .
p. w. anderson , science * 235 * , 1196 ( 1987 ) .
f.c . zhang and t.m .
rice , phys .
* b37 * , 3759 ( 1988 ) .
e. dagotto , j. riera and d.j .
scalapino , phys .
45 , 5744 ( 1992 ) .
h. tsunetsugu , m. troyer and t.m .
rice , phys . rev . *
b49 * , 16078 ( 1994 ) .
m. troyer , h. tsunetsugu and t.m .
rice , phys . rev . * b53*,251 ( 1996 ) .
d. poilblanc , d , j .
scalapino and w. hanke , phys . rev . * b52 * , 6796 ( 1995 ) .
hayward and d. poilblanc , phys .
b53 , 11721 ( 1996 ) .
hayward , d. poilblanc and d.j .
scalapino , phys .
* b53 * , r8863 ( 1996 ) .
m " uller and t.m .
rice , phys .
b58 , 3425 ( 1998 ) . j. riera , d. poilblanc and e. dagotto , eur .
j. b7 , 53 ( 1999 ) .
m. brunner , s. capponi , f.f .
assaad and a. muramatsu , phys . rev .
* b63 * , 180511 ( 2001 ) .
s.r . white and d.j .
scalapino , phys .
* b55 * , 6504 ( 1997 ) .
g. sierra , m.a .
martin - delgado , j. dukelsky , s.r .
white and d.j .
scalapino , phys . rev . *
b57 * , 11666 ( 1998 ) .
s.r . white and d.j .
scalapino , phys . rev . *
b57 * , 11666 ( 1998 ) .
s. rommer , s.r .
white and d.j .
scalapino , phys . rev . * b61 * , 13424 ( 2000 ) . t. siller , m. troyer , t.m . rice and s.r .
white , phys .
b63 , 195106 ( 2001 ) .
white , i. affleck and d.j .
scalapino , phys .
b65 , 165122 ( 2002 ) .
u. schollwock , s. chakravarty , j.o .
fjaerstad , j.b . marston and m. troyer ,
90 , 186401 ( 2003 ) .
d. poilblanc , e. orignac , s.r .
white and s. capponi , phys .
b69 , 220406(r ) ( 2004 ) .
sushkov , phys .
b60 , 3289 ( 1999 ) .
m. sigrist , t.m .
rice and f.c .
zhang , phys .
b49 , 12058 ( 1994 ) .
lee , y.w .
lee , c.y .
mou and z.y .
weng , phys . rev . *
b60 * , 13418 ( 1999 ) .
zheng and c.j .
hamer , phys .
b66 , 085112 ( 2002 ) .
j. oitmaa , c.j . hamer and zheng w - h .
b60 , 16364 ( 1999 ) . c. jurecka and w. brenig , physica b - cond
. mat . * 312 * , 592 ( 2002 ) .
batista and g. ortiz , phys .
letts . * 85 * , 4755 ( 2000 ) .
t. barnes , e. dagotto , a. moreo and e.s .
swanson , phys .
b40 , 10977 ( 1989 ) .
j. riera and e. dagotto , phys . rev .
* b47 * , 15346 ( 1993 ) .
a.l . chernyshev and p.w .
leung , phys .
b60 , 1592 ( 1999 ) .
riera , phys .
b64 , 104520 ( 2001 ) .
y. nagaoka , phys .
147 , 392 ( 1966 ) .
w.f . brinkman and t.m .
rice , phys . rev .
b2 , 1324 ( 1970 ) .
b.i . shraiman and e.d .
siggia , phys .
60 , 740 ( 1988 ) .
kane , p.a .
lee and n. read , phys .
b39 , 6880 ( 1989 ) .
white and i. affleck , phys
b64 , 02411 ( 2001 ) .
j. schulz , phys .
rev . b*59 * , 2471 ( 1999 ) .
s. r. white , phys .
* 69 * , 2863 ( 1992 ) ; phys.rev .
b * 48 * , 10345 ( 1993 ) ; i. peschel , x. wang , m. kaulke and k. hallberg ( eds ) , _ density matrix renormalization _ , springer ( 1999 ) . g. martinez and p. horsch , phys
* b44 * , 317 ( 1991 ) .
z. liu and e. manousakis , phys . rev . * b45 * , 2425 ( 1992 ) .
hamer , zheng w - h . and
j. oitmaa , phys .
rev . * 58 * , 15508 ( 1998 ) .
s. schmitt - rink , c.m .
varma and a.e .
ruckenstein , phys .
* 60 * , 2793 ( 1988 ) .
m. ogata , m. u. luchini , s. sorella , f.f .
assaad , phys .
66 * , 2388 ( 1991 ) . | the phase diagram of the two - leg t-@xmath0 ladder is explored , using the density matrix renormalization group method .
results are obtained for energy gaps , electron density profiles and correlation functions for the half - filled and quarter - filled cases .
the effective lagrangian velocity parameter @xmath1 is shown to vanish at half - filling .
the behaviour of the one - hole gap in the nagaoka limit is investigated , and found to disagree with theoretical predictions .
a tentative phase diagram is presented , which is quite similar to the full t - j ladder , but scaled up by a factor of about two in coupling .
near half - filling a luther - emery phase is found , which may be expected to show superconducting correlations , while near quarter - filling the system appears to be in a tomonaga - luttinger phase . |
theoretical studies@xcite predict that the magnetic geometrically nonfrustrated diamond lattice can host a highly degenerate frustrated state consisting of coplanar spirals , the so - called spiral spin liquid ( ssl ) .
this counterintuitive finding is comprehended when taking into account that the diamond lattice is composed of two interpenetrated face - centered cubic ( fcc ) sublattices which on their own are highly frustrated@xcite .
the magnetic frustration of the overall diamond lattice is caused by a significant j@xmath0 interaction which couples nearest - neighbor sites within each fcc sublattice .
the ssl should evolve@xcite when the ratio @xmath4/@xmath3 exceeds @xmath8 , while for lower @xmath4/@xmath3 a collinear antiferromagnet ( af ) should be the ground state .
the labeling of j s reflects the fact that j@xmath9 is the nearest - neighbor coupling of the diamond lattice and j@xmath0 is the next - nearest - neighbor one .
the degeneracy of the ssl could be lifted by thermal@xcite or quantum@xcite fluctuations resulting in a magnetic ordering transition via the order - by - disorder mechanism@xcite . within this scenario
the ordered state selected by entropy has the highest density of nearby low - energy excitations .
this model will be referred to as the @xmath4/@xmath3 or order - by - disorder model .
+ coal@xmath0o@xmath1 is a good candidate to study ssl , as it adopts the spinel structure ab@xmath0o@xmath1 with magnetic co@xmath2 ions residing on the diamond lattice .
however , there is no consensus to which extent the @xmath4/@xmath3 model is relevant for this system .
the ratio @xmath4/@xmath3 determined from inelastic neutron scattering@xcite is only 0.109(2 ) , thus , according to the aforementioned model@xcite , this material should have a nondegenerate collinear af ground state
. nevertheless , experimental observations by specific heat@xcite , muon spin relaxation@xcite , neutron diffraction@xcite are rather consistent with an unconventional ground state , which might be the ssl .
several models have been proposed in the past to explain this behaviour , such as the order - by - disorder @xcite , an order - by - quenched - disorder @xcite or a kinetically inhibited order @xcite models .
+ in this work we investigate the relevance of an order - by - quenched - disorder model to our system .
it is focused on the well - known experimental fact that materials with the spinel structure could have a significant degree of an a / b antisite inversion .
that is , b - atoms occupy the a - site and a - atoms - the b - site . in coal@xmath0o@xmath1 samples studied so far@xcite the co
/ al inversion @xmath10 varies from 2 to 17% .
the theory@xcite predicts that impurities would break the degeneracy of the spiral spin liquid and would induce an ordered magnetic ground state , which is different from the one favoured by the order - by - disorder mechanism .
finally , the kinetically inhibited order model infers that in coal@xmath0o@xmath1 freezing of magnetic domain walls prevents a true long - range order , @xmath11 the thermal energy is too small to overcome the surface energy of domain walls .
thus the size of domains stays arrested at lowest temperatures even in the absence of antisite disorder .
+ for better understanding of the nonconventional magnetic order in the coal@xmath0o@xmath1 system we focus on neutron single crystal diffraction which probes spin - spin correlations and thus should be able to distinguish the models introduced above .
we calculate diffraction patterns for magnetic moment arrangements obtained by the classical monte - carlo ( mc ) simulations varying the @xmath4/@xmath3 ratio , magnetic field and disorder on the a- and b- sites and compare those with experimental patterns .
measurements of susceptibility , magnetization and specific heat complement our neutron diffraction results .
we conclude that both , the frustration due to the @xmath4/@xmath3 ratio and the presence of the antisite disorder , play a significant role in the observed atypical magnetic ordering .
we anticipate that disorder is not random , presumably it happens at @xmath6111@xmath7-twin boundaries restricting the propagation of the collinear af state .
[ sec2 ] the monte - carlo calculation was performed minimizing the energy of the classical heisenberg hamiltonian for spins @xmath12 : @xmath13 /@xmath3=0.10 ( left ) , 0.17 ( right ) .
middle : the 00l cuts of the h - hl maps with the @xmath4/@xmath3 ratio 0.10 ( red ) , 0.15 ( green ) and 0.17 ( blue ) . note the logarithmic scale for the intensity.,title="fig : " ] /@xmath3=0.10 ( left ) , 0.17 ( right ) .
middle : the 00l cuts of the h - hl maps with the @xmath4/@xmath3 ratio 0.10 ( red ) , 0.15 ( green ) and 0.17 ( blue ) . note the logarithmic scale for the intensity.,title="fig : " ] /@xmath3=0.10 ( left ) , 0.17 ( right ) .
middle : the 00l cuts of the h - hl maps with the @xmath4/@xmath3 ratio 0.10 ( red ) , 0.15 ( green ) and 0.17 ( blue ) . note the logarithmic scale for the intensity.,title="fig : " ] here @xmath14 are first- and second - neighbour antiferromagnetic couplings , @xmath15 and @xmath16 are the magnitude and direction of the single - ion anisotropy term , @xmath17 and @xmath18 are the magnitude and direction of a magnetic field . a cluster of 64000 co atoms ( 20x20x20 unit cells ) with periodic boundary conditions was considered .
this size is a compromise between two conflicting requirements - the cluster should be sufficiently large to ensure enough resolution in a calculated diffraction pattern , but small enough to spend finite time on a mc calculation .
the moments were kept constant in magnitude and equal to 1 @xmath19 ; their direction was changed at random , one at the time . to obtain the ground state only energy - decreasing moves were accepted in the final stage . during a mc cycle 10000 moves of each atom
were performed to ensure equilibrium conditions .
the results were identical to those obtained by temperature annealing .
the fourier transform of the magnetic moment configuration was then calculated for the regions of interest of the reciprocal space .
the q - step was chosen to be 0.05 r.l.u .
no binning in the direction perpendicular to the cuts was performed .
+ as the main purpose of the modelling was to understand the magnetic behaviour in the coal@xmath0o@xmath1 system , the exchange couplings and disorder parameter close to ones known for our crystal @xcite were used as the basis , namely @xmath3=9.2 k , @xmath4/@xmath3=0.10 , d / j@xmath9=0.009 and @xmath10=8% @xcite . additionally , to disentangle the influence of the parameters , we varied in reasonable ranges the @xmath4/@xmath3 ratio , amount of random defects and magnetic field strength .
we tried two different single - ion anisotropy directions - @xmath20111@xmath21 $ ] and @xmath20100@xmath21 $ ] , however , as the anisotropy in this system is small this did not influence the results . /@xmath3=0.10 : 0% ( black ) , 4% ( green ) , 8% ( blue ) , 12% ( red ) .
note the logarithmic scale for the intensity.,title="fig : " ] /@xmath3=0.10 : 0% ( black ) , 4% ( green ) , 8% ( blue ) , 12% ( red ) .
note the logarithmic scale for the intensity.,title="fig : " ] + representative h@xmath22hl maps for the ideal ( without defects ) model with @xmath4/@xmath3 ratios of 0.10 and 0.17 are presented in fig .
two main features are apparent : there are sharp magnetic bragg peaks at the af ( k=0 ) positions ( i.e. 200 , 111 ) and broad diffuse features at the basis of magnetic bragg peaks extending over several reciprocal lattice units ( r.l.u . ) .
the broad diffuse features need the log - scale amplification in single crystal calculated patterns , however summing up this sparse signal at the same sin@xmath23 results in a strong diffuse scattering in powder patterns@xcite . as can be seen from the 00l cuts in fig .
[ fig1 ] ( middle ) , the broad diffuse features grow with the increase of @xmath4/@xmath3 ( a significant increase happens for @xmath4/@xmath24 0.15 ) , while the bragg intensity decreases and even a hollow forms at the positions of the magnetic bragg peaks for @xmath4/@xmath3= 0.17 .
this reproduces well the spiral surfaces of the free energy calculated by bergman _
[ ] . + /@xmath3= 0.10 model with magnetic fields applied along @xmath20110@xmath21 $ ] ( left ) and @xmath20100@xmath21 $ ] ( middle ) and for the model with 8% defects for 4 t with field along @xmath20100@xmath21 $ ] ( red ) and @xmath20110@xmath21 $ ] ( green ) ( right).,title="fig : " ] /@xmath3= 0.10 model with magnetic fields applied along @xmath20110@xmath21 $ ] ( left ) and @xmath20100@xmath21 $ ] ( middle ) and for the model with 8% defects for 4 t with field along @xmath20100@xmath21 $ ] ( red ) and @xmath20110@xmath21 $ ] ( green ) ( right).,title="fig : " ] /@xmath3= 0.10 model with magnetic fields applied along @xmath20110@xmath21 $ ] ( left ) and @xmath20100@xmath21 $ ] ( middle ) and for the model with 8% defects for 4 t with field along @xmath20100@xmath21 $ ] ( red ) and @xmath20110@xmath21 $ ] ( green ) ( right).,title="fig : " ] to understand the effect of the co / al antisite disorder we considered random defects on the two sites a ( co)- and b ( al ) separately .
we considered models with random a - defects and b - defects .
first , up to 12% of magnetic atoms were removed from the a - sites .
this significantly changed the intensity distribution , as is presented in fig .
[ fig2 ] left for the ratio @xmath4/@xmath3=0.10 .
there appeared sharp ferromagnetic peaks ( _ i.e. _ 000 , 220 ) , the af bragg peaks became wider and the broad diffuse scattering increased .
these tendencies grow with the number of the a - defects .
inspection of atoms in the vicinity of the defects shows that the antiferromagnetic coupling between them is broken and their magnetic moments are aligned along the single - ion anisotropy axis @xmath16 .
this results in the emergence of long - range ferromagnetic correlations at the cost of the antiferromagnetic ones .
+ next we inserted up to 12% of magnetic atoms on the b - sites ( see fig .
[ fig2 ] right ) .
this model , introduced first by savary_et al . _
[ ] , argues that the coupling between the magnetic b - atom and its six nearest - neighbor a - atoms should be very strong .
we adopted this model to the coal@xmath0o@xmath1 case considering the ratio @xmath4/@xmath3=0.10 and we varied the concentration of impurities in the 2 - 12% range .
the results imply that b - defects suppress the af long - range order much stronger than the a - defects . for concentrations above 4% antiferromagnetic bragg peaks vanish .
also at these concentrations ferromagnetic short - range correlations emerge ; they were not detected either for the ideal or a - defect models .
/@xmath3=0.10 with a magnetic field applied along @xmath20100@xmath21$].,title="fig : " ] /@xmath3=0.10 with a magnetic field applied along @xmath20100@xmath21$].,title="fig : " ] + further , we wanted to know how a perturbation , such as a magnetic field , would affect diffraction patterns of the ideal and defect models . for
that we calculated diffraction patterns for the @xmath4/@xmath3=0.10 ratio without defects and with a - defects .
application of magnetic fields along the principal directions lead to significant changes in magnetic correlations for the ideal model ( fig .
[ fig3 ] left and middle ) .
the common features are : increase of ferromagnetic bragg peaks , decrease of antiferromagnetic peaks ( they vanish at the magnetic field of 5 tesla ) and enhancement of broad diffuse af features .
still , there are differences .
when magnetic field is applied along @xmath20110@xmath21 $ ] ( fig .
[ fig3 ] left ) , diffuse scattering is concentrated around the bragg peaks , the peaks themselves stay sharp . for h@xmath25100@xmath21
$ ] ( fig .
[ fig3 ] middle ) the af peaks become wider . in addition to diffuse scattering located near the af bragg peaks ( as presented in fig .
[ fig4 ] left for the 0kl plane with h@xmath25100@xmath21 $ ] ) , short - range correlations form diffuse streaks along @xmath6111@xmath7 .
they are more pronounced in the @xmath6hk - k@xmath7 plane for h@xmath25100@xmath21 $ ] ( fig .
[ fig4 ] right ) compared to the @xmath6h - hl@xmath7 plane for h@xmath25110@xmath21 $ ] ( not shown ) .
this suggests that magnetic field along @xmath20100@xmath21 $ ] lowers the energy of low - lying states more efficiently than magnetic field along @xmath20110@xmath21 $ ] . in clear contrast , when 8% a - defects are present ( fig .
[ fig3 ] right ) , the magnetic field enhances all magnetic correlations - not only the ferromagnetic long - range bragg contributions and af diffuse scattering , but also the af bragg contributions grow .
+ our mc calculations show that magnetic field is a sensitive parameter to identify the relevance of the @xmath4/@xmath3 model and to distinguish the presence or absence of random a- and b - defects .
following these conjectures we decided to perform a neutron single crystal diffraction experiment in the applied magnetic field .
[ sec3 ] prior to neutron diffraction we characterized our single crystal by bulk measurements to compare our results with other numerous studies of the coal@xmath0o@xmath1@xcite .
the growth of the coal@xmath0o@xmath1 crystal by the floating zone method was reported by maljuk _
. _ @xcite ; parts of this crystal were used in the previous neutron scattering experiments @xcite .
susceptibility , magnetization and specific heat measurements presented here were performed on a part of this crystal with the mass of 12.5 mg using a quantum design squid vibrating sample magnetometer ( svsm ) and a quantum design ppms , respectively .
+ measured with a dc field @xmath26=0.005 t. solid symbols represent field cooled and open symbols - zero field cooled data .
green - magnetic field applied along the @xmath20111@xmath21 $ ] direction , blue - @xmath20110@xmath21 $ ] , red - @xmath20100@xmath21 $ ] .
inset : inverse susceptibility for three principal directions .
middle : temperature variation of ac magnetic susceptibility @xmath27 for different frequencies with a dc field @xmath26= 0.1 t along @xmath20111@xmath21 $ ] .
right : magnetization at 2 k ( open symbols ) and 30 k ( solid symbols ) . insets : a ) change of @xmath28 with h@xmath29111@xmath21 $ ] .
b ) variation of @xmath28 with temperature at different strength of magnetic field along @xmath20111@xmath21$].,title="fig : " ] measured with a dc field @xmath26=0.005 t. solid symbols represent field cooled and open symbols - zero field cooled data .
green - magnetic field applied along the @xmath20111@xmath21 $ ] direction , blue - @xmath20110@xmath21 $ ] , red - @xmath20100@xmath21 $ ] .
inset : inverse susceptibility for three principal directions .
middle : temperature variation of ac magnetic susceptibility @xmath27 for different frequencies with a dc field @xmath26= 0.1 t along @xmath20111@xmath21 $ ] .
right : magnetization at 2 k ( open symbols ) and 30 k ( solid symbols ) .
insets : a ) change of @xmath28 with h@xmath29111@xmath21 $ ] .
b ) variation of @xmath28 with temperature at different strength of magnetic field along @xmath20111@xmath21$].,title="fig : " ] measured with a dc field @xmath26=0.005 t. solid symbols represent field cooled and open symbols - zero field cooled data . green - magnetic field applied along the @xmath20111@xmath21 $ ] direction , blue - @xmath20110@xmath21 $ ] , red - @xmath20100@xmath21 $ ] .
inset : inverse susceptibility for three principal directions .
middle : temperature variation of ac magnetic susceptibility @xmath27 for different frequencies with a dc field @xmath26= 0.1 t along @xmath20111@xmath21 $ ] .
right : magnetization at 2 k ( open symbols ) and 30 k ( solid symbols ) . insets : a ) change of @xmath28 with h@xmath29111@xmath21 $ ] .
b ) variation of @xmath28 with temperature at different strength of magnetic field along @xmath20111@xmath21$].,title="fig : " ] the low temperature zero field cooled ( zfc ) and field cooled ( fc ) susceptibility measured with a dc magnetic field @xmath26= 0.005 t applied along three principal directions is shown in fig .
[ fig5 ] left .
the high - temperature part , above 100 k , follows a curie - weiss law with a weiss temperature @xmath30= -85.2 k. the determined effective magnetic moment 4.60 @xmath19/f.u . and
g - factor 2.37 are in good agreement with other studies @xcite .
a broad hump at @xmath3111 k indicates building up of short range antiferromagnetic correlations .
below t@xmath328 k a small splitting between the zfc and fc susceptibilities without a pronounced cusp is observed .
remarkably , there is no frequency dependence of this anomaly in the ac susceptibility ( fig .
[ fig5 ] middle ) .
these observations are in excellent agreement with the systematic study of antisite disorder in the co@xmath33al@xmath34(al@xmath33co@xmath35)@xmath0o@xmath1 system by hanashima _
et al._@xcite , which attributes such behaviour to a spin liquid state ( samples of suzuki _
et al._@xcite and roy _
et al._@xcite belong to this regime as well ) .
et al._@xcite report a crossover between spin liquid ( sl ) and spin glass ( sg ) regimes with a threshold at the 8% inversion level .
samples belonging to the sg regime have a cusp in the dc susceptibility at the zfc and fc splitting point and the real part of the ac susceptibility is frequency dependent ( samples of tristan
_ et al . _
@xcite belong to this range ) . presumably in the sl regime
defects only perturb the degenerate ground state , while in the sg regime they start to dominate magnetic behaviour .
as both , the zfc and fc splitting in the susceptibility and the broad magnetic peaks in neutron diffraction in our crystal @xcite , appear at t@xmath368 k , we identify this anomaly as the emergence of an unconventional ordered magnetic state .
+ isothermal magnetization curves for our crystal with the field along three principal axes for 2 k and 30 k are presented in fig .
[ fig5 ] right . at 30 k magnetization
is linear with the field as expected for a paramagnet , while at 2 k the behavior is nonlinear ( as emphasized in the inset ( a ) for @xmath20111@xmath21 $ ] direction ) .
such nonlinearity would be expected for reorientation of magnetic domains , however @xmath28 is identical regardless of the field direction thus suggesting that magnetic anisotropy in this system is very weak .
+ measured in zero and 6 t magnetic field applied along three principal directions : red @xmath20100@xmath21 $ ] , blue @xmath20110@xmath21 $ ] , green @xmath20111@xmath21 $ ] .
heat capacity of znal@xmath37o@xmath38 ( non - magnetic analog ) taken from ref .
[ ] is also plotted .
inset : difference of c@xmath39 measured in 6 t and without field .
right : magnetic entropy @xmath40 calculated from these data .
the dashed horizontal line indicates the high temperature limit of @xmath40= 11.56 j / mol k for s=3/2 .
inset : magnetic contribution to the heat capacity c@xmath41 fitted to a power law with @xmath42=2.1.,title="fig : " ] measured in zero and 6 t magnetic field applied along three principal directions : red @xmath20100@xmath21 $ ] , blue @xmath20110@xmath21 $ ] , green @xmath20111@xmath21 $ ] .
heat capacity of znal@xmath37o@xmath38 ( non - magnetic analog ) taken from ref .
[ ] is also plotted .
inset : difference of c@xmath39 measured in 6 t and without field .
right : magnetic entropy @xmath40 calculated from these data .
the dashed horizontal line indicates the high temperature limit of @xmath40= 11.56 j / mol k for s=3/2 .
inset : magnetic contribution to the heat capacity c@xmath41 fitted to a power law with @xmath42=2.1.,title="fig : " ] the temperature dependence of the heat capacity measured in zero field and in the applied magnetic field of 6 t along the three principal directions are very much alike ( fig .
[ fig6 ] left ) . for all measurements the crystal
was cooled in zero field and then a magnetic field was applied . a broad peak located at @xmath319.8
k in zero field slightly shifts to lower temperature when magnetic field is applied ( inset of fig .
[ fig6 ] left ) .
the change though tiny and independent of the field direction , suggests that only few degrees of freedom are involved and their response to the magnetic field averages throughout the sample . to obtain the magnetic contribution
@xmath43 we subtracted the heat capacity of the crystal lattice modelled by the debye approximation @xcite . as shown in the inset of fig .
[ fig6 ] right , @xmath43 follows the power law t@xmath44 with @xmath42=2.1 ( red solid line ) within the temperature range @xmath45 k. in other samples reported in literature the power laws vary from @xmath42=2.5 , @xmath42=2.23 for the sl regime samples @xcite to @xmath42=2 for the gs sample @xcite . these @xmath42-values should be compared to the power law expected for a classical antiferromagnetic order ( @xmath46 t@xmath47 ) , for the spiral surface ( @xmath4/@xmath3 ) model ( @xmath46 t@xmath48)@xcite and a linear behavior typical for spin glass systems ( @xmath46 t ) .
the agreement is in favor of the @xmath4/@xmath3 model and the presence of defects seems to affect the entropy very little .
notably , the saturation of the magnetic entropy @xmath40 evaluated by integrating @xmath49 over t ( fig .
[ fig6 ] right ) is significantly lower than the expected high - temperature limit for @xmath50 systems ( 11.56 j / molk ) .
presumably a significant part of entropy remains in the system up to higher temperatures .
neutron single crystal experiments were performed on a 0.2755 g crystal cut from the coal@xmath0o@xmath1 ingot at the base temperature of 1.8 k on the single crystal diffractometer trics at sinq , villigen .
magnetic field was applied vertically along the principal directions @xmath20010@xmath21 $ ] , @xmath20110@xmath21 $ ] and @xmath20111@xmath21 $ ] by reorienting the crystal in a ma06 cryomagnet .
the neutron wavelength of 1.18 and a 80 collimator in front of a @xmath51he detector were used . in zero - field ,
the magnetic peaks of the @xmath600@xmath52 family are significantly ( _ c.a .
_ 3 times ) broaden ( fig .
[ fig7 ] ) compared to the instrumental resolution .
this resembles broad peaks of the mc model with a - defects and not the features of the ideal mc model ( sharp peaks and broad diffuse scattering extending over several r.l.u . ) .
however , contrary to the predictions of the models with random defects , no magnetic intensity is present at the ferromagnetic ( nuclear ) positions in zero field .
+ in a magnetic field , the intensity and shape of magnetic peaks change and these changes strongly depend on the field direction .
the magnetic field applied along the @xmath20110@xmath21 $ ] direction ( fig .
[ fig7 ] left ) reduces the width of the ( 002 ) magnetic reflection and significantly ( 15% ) increases the peak intensity at the maximum .
however , the total intensity remains almost the same .
this observation suggests that long - range correlations grow at the cost of short - range correlations perpendicular to ( 002 ) being in accord with the increase of the af long - range correlations in the mc model with a - defects .
+ for h@xmath29010@xmath21 $ ] ( fig .
[ fig7 ] right ) the intensity on the ( 200 ) magnetic peak maximum decreases by 37% and the total intensity decreases by 21% .
this contradicts the mc model with a - defects , which predicts an enhancement of the af long - range correlations .
the field applied along the @xmath20111@xmath21 $ ] direction ( fig .
[ fig7 ] middle ) changes the peak maximum and the total intensity of the magnetic peaks ( 200 ) , ( 00 - 2 ) and ( 020 ) very little .
this also contradicts the mc model with a - defects .
+ it should be noted that our experimental findings can not be explained within the conventional model of a multidomain af structure either .
that model postulates the presence of magnetic domains , each of them having the same magnetic structure with a fixed magnetic moment direction along one of complementary preferred easy axes ( for example , octet of the @xmath6111@xmath7 axes ) .
when a magnetic field is applied , domains with magnetic moments orthogonal to the field direction grow at the cost of domains with unfavourable moment axes . in our case
, however , it seems that magnetic moments rotate orthogonal to the applied magnetic field but not necessarily into domains of the complementary easy axis .
such soft magnetic behaviour can explain the increase of the ( 002 ) intensity maximum for h@xmath25110@xmath21 $ ] and its reduction for the h@xmath25010@xmath21 $ ] magnetic field direction .
+ 200@xmath7 reflections of coal@xmath37o@xmath38 crystal measured at 1.8k without ( 0 t , black ) and with ( 5 t , red ) magnetic field applied along @xmath20110@xmath21 $ ] ( left ) , @xmath20111@xmath21 $ ] ( middle ) and @xmath20010@xmath21 $ ] ( right).,title="fig : " ] 200@xmath7 reflections of coal@xmath37o@xmath38 crystal measured at 1.8k without ( 0 t , black ) and with ( 5 t , red ) magnetic field applied along @xmath20110@xmath21 $ ] ( left ) , @xmath20111@xmath21 $ ] ( middle ) and @xmath20010@xmath21 $ ] ( right).,title="fig : " ] 200@xmath7 reflections of coal@xmath37o@xmath38 crystal measured at 1.8k without ( 0 t , black ) and with ( 5 t , red ) magnetic field applied along @xmath20110@xmath21 $ ] ( left ) , @xmath20111@xmath21 $ ] ( middle ) and @xmath20010@xmath21 $ ] ( right).,title="fig : " ]
[ sec4 ] in sec . [ sec2 ]
we calculated the single - crystal neutron diffraction patterns based on the ideal order - by - disorder model and extended the simulation to the order - by - quenched - disorder models with random a- and b- defects . in sec .
[ sec3 ] our experimental data on our coal@xmath0o@xmath1 single crystal are presented .
our macroscopic ( susceptibility , magnetization and specific heat ) and microscopic ( single crystal neutron diffraction in zero and in the applied magnetic field ) experimental findings confirm that the magnetic order in the frustrated diamond lattice antiferromagnet coal@xmath0o@xmath1 is unconventional .
comparison of bulk properties of our crystal with those of other studied samples identifies our crystal as located in the spin liquid regime@xcite and its magnetic properties are isotropic on the macroscopic scale .
this is definitely not the case on the microscopic level as testified by neutron diffraction .
single crystal patterns reveal that changes of magnetic correlations with the applied field are anisotropic .
the patterns are similar to those calculated from the monte - carlo models , but none of the considered models explains the data completely . in zero field , the measured peaks are broadened , as the model with a - defects predicts , however , the anticipated ferromagnetic contribution is missing .
in addition , while the experimentally observed increase of long - range af correlations for h@xmath25110@xmath21 $ ] is in accord with this model , its decrease for h@xmath25100@xmath21 $ ] contradicts the predictions .
furthermore , the presence of random b - defects should induce broad ferromagnetic features , but they are not observed in the experiment .
+ thus , the magnetic behaviour in the studied material is not captured by simplistic models supposing random disorder on the a- and b - sites considered so far .
one of the possible reasons of the failure could be different electronic states of the defect atoms as the assumed ones .
for example , the co@xmath53 ions inserted into the octahedral b - sites are assumed to be in the same spin state as in the a - site .
however dft calculations @xcite suggest the low spin configuration for the co@xmath53 ions in the octahedral coordination for the inversion levels up to @xmath10=50% .
another possible reason of disagreement could be a complex morphology of defects which we assumed to be random .
spinels are known to form complicated twin structures @xcite , which lead to stacking faults of the @xmath6111@xmath7 planes .
such faults have been observed in mgal@xmath0o@xmath1 by transmission electron microscopy ( tem ) @xcite .
starting from the ideal spinel structure shown in fig .
[ fig8 ] ( a ) a stacking fault develop at the ( 111 ) plane indicated by the black solid line . at such twin boundary
the cation composition can be either preserved ( fig .
[ fig8 ] b ) , or it can become b(al)-rich ( fig .
[ fig8 ] c ) or a(co)-rich ( fig . [ fig8 ] d ) .
if the distribution of stacking faults is similar in all eight directions within the @xmath6111@xmath7 family , macroscopic magnetic properties of a crystal would remain also isotropic .
+ clearly , the frustration due to the @xmath4/@xmath3 ratio is the basic ingredient of the magnetic behaviour of coal@xmath0o@xmath1 .
this is the case even when the @xmath4/@xmath3 ratio is below @xmath8 .
the @xmath4/@xmath3 model correctly predicts the temperature evolution of specific heat , the wave vector of magnetic ordering and the presence of low - lying excited states with @xmath54= @xmath6111@xmath7 .
however , not all experimental observations emerge from the @xmath4/@xmath3 physics . both , the difference in zfc / fc susceptibility and the broadening of the af peaks in neutron diffraction result from the antisite disorder .
comparison of neutron patterns from the mc modelling and the experiment prompts that these models are too simplistic .
tem would be the key method for a systematic study of the microstructure and its influence on magnetic correlations in this material reconciling contradicting isotropic macroscopic and anisotropic microscopic observations .
+ 1 - 10@xmath21 $ ] projection of the ab@xmath0o@xmath1 spinel structure with a ) ideal arrangement of co@xmath53 a - tetrahedra ( violete ) and al@xmath55 b - octahedra ( green ) and @xmath6111@xmath7-twin boundaries preserving the cation composition ( b ) or resulting in the b - rich ( c ) , a - rich ( d ) regions . ]
99 d. bergman , j. alicea , e. gull , _
et al . _ , nature physics * 3 * , 487 ( 2007 ) .
m. v. gvozdikova , m. e. zhitomirsky , jetp lett .
* 81 * , 236 ( 2005 ) .
bernier , m. j. lawler , y. b. kim , phys .
* 101 * , 047201 ( 2008 ) .
j. villian , r. bidaux , j .-
carton , r. conte , j. phys . * 41 * , 1263 ( 1980 ) . c. l. henley , phys .
rev . lett . *
62 * , 2056 ( 1989 ) .
n. tristan , j. hemberger , a. krimmel , h .- a .
krug von nidda , v. tsurkan , and a. loidl , phys .
b * 72 * , 174404 ( 2005 ) .
n. tristan , v. zestrea , g. behr , _ et al .
b * 77 * , 094412 ( 2008 ) .
t. suzuki , n. nagai , m. nohara , h. takagi , j. phys . : condens .
matter * 19 * , 145265 ( 2007 ) .
o. zaharko , a. cervellino , v. tsurkan , n. b. christensen , a. loidl , phys .
b * 81 * , 064416 ( 2010 ) .
o. zaharko , n. b. christensen , a. cervellino , v. tsurkan , a. maljuk , u. stuhr , c. niedermayer , f. yokaichiya , d. n. argyriou , m. boehm , a. loidl , phys .
b * 84 * , 094403 ( 2011 ) .
g. j. macdougall , d. gout , j. l. zarestky , g. ehlers , a. podlesnyak , m. a. guire , d. mandrus , s. e. nagler , pnas * 108 * , 15693 ( 2011 ) .
l. savary , e. gull , s. trebst , j. alicea , d. bergman , l. balents , phys .
b * 84 * , 064438 ( 2011 ) .
k. hanashima , y. kodama , d. akahoshi , _ et al .
_ , j. phys .
. jpn . * 82 * , 024702 ( 2013 ) . preparing this manuscript we revisited the x - ray diffraction refinements and found that the degree of inversion in our crystal is actually lower , 5% .
a. maljuk , v. tsurkan , v. zestrea , o. zaharko , a. cervellino , a. loidl , d. n. argyriou , j. crystal growth * 311 * , 3997 ( 2009 ) .
b. roy , abhishek pandey , q. zhang , t. w. heitmann , d. vaknin , d. c. johnston , and y. furukawa , phys .
b * 88 * , 174415 ( 2013 ) .
p. debye , ann .
( leipzig ) * 39 * , 789 ( 1912 ) .
f. tielens , m. calatayud , r. franco , j. m. recio , j. perez - ramire , c. minot , j. phys .
chem . * 110 * , 988 ( 2006 ) . c. b. carter , z. elgat , t. m. shaw , phil . mag .
a * 55 * , 1 ( 1987 ) .
h. klapper , th .
hahn , acta cryst a * 68 * , 82 ( 2012 ) . | coal@xmath0o@xmath1 spinel with magnetic co@xmath2 ions on the diamond a - lattice is known to be magnetically frustrated .
we compare neutron single crystal diffraction patterns measured in zero and applied magnetic fields with the ones obtained from classical monte - carlo models . in simulations we test the influence of various parameters on diffraction patterns : the ratio of nearest- , @xmath3 , and next - nearest , @xmath4 , neighbor interactions , magnetic field applied along the principal crystallographic directions , and random disorder on the a(co@xmath2)- and b(al@xmath5)- sites .
we conclude that the models considered so far explain the broadening of magnetic bragg peaks in zero magnetic field and their anisotropic response to applied magnetic field only partly . as bulk properties of our single crystal
are isotropic , we suggest that its microstructure , specifically @xmath6111@xmath7-twin boundaries , could be a reason of the nonconventional magnetic order in coal@xmath0o@xmath1 . |
in spite of factors such as a strong tidal field due to the central supermassive black hole sgr a * ( e.g. , * ? ? ? * ; * ? ? ?
* ) , or strong stellar winds , which pose serious obstacles for star formation , surprisingly , numerous young ( a few times @xmath3yrs ) , massive stars can be found in the central parsec , e.g. , the one to two dozen bright so - called he - stars .
the young , massive stars are mainly concentrated in the region @xmath010@xmath4 in projection around sgr a * and appear to rotate around sgr a * in two counter - rotating disks that contain an apparently coeval population of wolf - rayet ( wr ) , luminous blue variables ( lbv ) , and o / b - stars ( * ? ? ?
* ; * ? ? ?
* also paumard et al .
2005 , in preparation ) .
also , there are b - type main sequence stars in the immediate vicinity of sgr a * ( e.g. , * ? ? ?
* ; * ? ? ?
the presence of the young , massive stars near sgr a * is not well understood and various explanations for their existence are currently being discussed , such as , e.g. , infall and collision of molecular clouds , infall and dissolution of a massive cluster ( e.g. , see discussions in * ? ? ?
* ) , or star formation in a self - gravitating accretion disk ( e.g. , * ? ? ?
* ; * ? ? ?
* ) . in this context ,
the irs 13e complex is a very intriguing object .
located @xmath5 in projection from sgr a * , it comprises almost half a dozen massive stars within a projected radius of @xmath00.25@xmath4 .
these stars appear to be wr stars or o - type supergiants .
in addition to representing an unusual concentration of bright , early - type stars , the irs 13e sources also have very similar proper motions ( see , e.g. , fig . 18 in * ? ? ? * ) . suggested that the irs 13e complex is the remnant core of a massive star cluster that had fallen into the central parsec and dissolved there .
their main arguments are a ) the similar proper motions of the irs 13e cluster members and b ) their finding that the stars in irs 13e appear to be massive , short - lived , and therefore young .
since one would expect that an association of stars such as irs 13e should be disrupted by the tidal field of sgr a * , they speculated that an intermediate - mass black hole ( imbh ) in irs 13e may prevent its disruption .
an imbh in the core of an infalling cluster would also provide an effective means of inward transportation through increased dynamical friction as has been suggested by @xcite ( see , however , * ? ? ?
the possible existence of an imbh in the gc is currently a hypothesis of high interest in the field
. therefore we consider it timely to present the newest data on stellar dynamics in irs 13e . in this letter , we analyze proper motions of the stars in and near the irs 13e complex , derived from ten years of near - infrared ( nir ) speckle and ao observations of the gc stellar cluster .
observations of the gc stellar cluster are routinely performed since spring 2002 with the conica / naos nir camera / adaptive optics ( ao ) system at the eso vlt unit telescope 4 on cerro paranal in chile . for this work ,
we used k - band ( and some h - band ) imaging data .
after standard data reduction ( sky subtraction , dead pixel correction , flat fielding , shift - and - add ) the final images were lucy - richardson ( lr ) deconvolved and beam restored with a gaussian beam of @xmath060mas ( 40mas for h - band images , respectively ) , corresponding to the diffraction limit of the vlt at @xmath6 m .
stellar positions were extracted with _
starfinder _ .
they were transformed into a coordinate system relative to sgr a * with the aid of 18 reference stars with well known positions and proper motions , taken from @xcite .
each of the data sets was divided into two parts that were reduced and analyzed independently .
uncertainties of the source positions were thus obtained by a comparison of two independent measurements for each epoch .
we added to these data gemini north ao h and k imaging data from july 2000 .
the images are part of the gemini north galactic center demonstration science data set .
both images were lr deconvolved and beam restored .
source positions were extracted in the way described above ( using the two independent measurements provided by the h and k images ) . in order to obtain a larger time baseline
, we also used sharp speckle imaging data from july 1995 , june 1996 , and june 2001 .
details on the observation and reduction of sharp imaging data may be found , e.g. , in @xcite .
again , we applied lr deconvolution and used two independent data sets for each epoch .
the irs 13e sources are named in fig.[fig : propmot ] .
@xcite label e3 and e4 as e3c and e3n , respectively . noted that e3 is a multiple source and named the two components e3a and e3b . on the image presented here
, e3 appears to have more than two components .
we just label the brightest one of them e3 ( the image in fig .
[ fig : propmot ] is based on data with a strehl @xmath7 in contrast to the gemini image of strehl @xmath05% used by ) . as concerns e5
, it appears highly confused with other sources and/or extended .
proper motions of the stars were determined by linear fits to the measured time - dependent positions with their respective uncertainties ( adopting a gc distance of 7.9 kpc from * ? ? ? * ) . in fig .
[ fig : propmot ] , the derived proper motions of all stars in and near irs 13e are shown superposed on an image .
the common proper motion of the irs 13e cluster members stands out clearly .
we identify five stars within a region of about @xmath8 diameter that could be part of irs 13e ( see also table [ tab : vels ] ) .
there may be further potential members , but they are either too weak , embedded in extended emission , or too close to brighter stars , such that no reliable proper motions could be determined . the proper motion of e6 ,
while pointing into the same direction as the ones of the other irs 13e sources , is significantly smaller in magnitude .
also , note that e6 is extincted much less than the other irs 13e stars .
it is therefore questionable whether e6 forms part of irs 13e and we omit it from the following analysis . the proper motion of e5 could not be reliably determined .
considered that irs 13e might be bound by an imbh . from spectroscopically
estimated radial velocities for e2 and e4 and assuming that the hypothetical imbh is located halfway between the two stars , they estimated its mass to @xmath01000m@xmath2 . with the proper motions , we can derive the minimum mass that is required to bind the irs 13e cluster members . here , it is first necessary to subtract the systemic motion of irs 13e . subsequently , the residual velocities of the individual sources can be used for estimating the mass in irs 13e under the assumption that it is a gravitationally bound system . since we can not know the true systemic velocity , we examined three cases : a ) using the average proper motion of the ensemble , b ) using the proper motion of e1 , and c ) using the proper motion of e4 . thus , all realistic cases , even extreme ones , should be covered . in the following ,
we limit our analysis to case a ) , which is the most conservative one because it results in the lowest enclosed masses . the average proper motion of irs 13e , derived from the mean of the velocities of sources e1 , e2 , e3 , and e4 , is @xmath9kms@xmath10 westwards and @xmath11kms@xmath10 northwards . here , we chose an unweighted average because we can not know which one of the individual proper motions corresponds best to the possible systemic velocity .
the proper motions of the irs 13e sources after subtraction of a systemic velocity corresponding to their mean motion are shown in the left panel of fig .
[ fig : masses ] .
a first rough estimate of the mass needed to bind irs 13e can be obtained from the residual velocities , through @xmath12 , where @xmath13 is the gravitational constant and @xmath14mpc the size of the system . with @xmath15kms@xmath10(@xmath16kms@xmath10 ) we obtain @xmath17m@xmath2(@xmath18m@xmath2 ) .
we chose two more refined approaches to estimating the mass of a putative imbh that gravitationally binds irs 13e : a ) estimating the mass in irs 13e with the leonard - merritt ( lm ) mass estimator ( eq . 19 in * ? ? ? * ) and b ) deriving a lower limit on the mass through the relation @xmath19 , with @xmath20 being the projected distance of a star from the imbh , @xmath21 its velocity projected onto the plane of the sky , @xmath22 its real distance from the imbh , @xmath23 its escape velocity , @xmath13 the gravitational constant , and @xmath24 the mass of the imbh . taking into account the unknown location of the black hole
, we examined a field of @xmath25 in right ascension and declination around the center of irs 13e with a grid step of @xmath26 .
the residual proper motion of each star results in a required lower mass for the imbh
dependent on its location in order to bind the star to irs 13e .
thus , one can obtain four maps for the sources e1 , e2 , e3 , and e4 .
the four maps were combined by taking the maximum ( otherwise the system would not be bound ) of the calculated mass limits at a given location .
we show the resulting map of lower mass limits in the right panel of fig .
[ fig : masses ] .
the significance of the calculated masses over the field is @xmath27@xmath28 .
since e1 has the proper motion with the strongest deviation from the mean proper motion of the ensemble , it is the star that gives the strongest constraints . there is a narrow minimum in the mass map for black hole locations close to e1 ( @xmath00.1@xmath4 nw of e1 ) .
in this most conservative case , the mass of the black hole would have to be larger than @xmath29m@xmath2 ( with a significance of 4@xmath28 ) .
if the black hole were located between e2 and e4 , the location used by for their mass estimate , it would have to have a mass @xmath30m@xmath2 ( significance @xmath31 ) .
these are only lower limits that rely on projected velocities and conservative estimates .
if irs 13e is indeed a bound system , a more realistic mass might be derived with the lm estimator .
again , there is a minimum near e1 for the black hole mass .
however , even in this conservative case , the lm estimator gives a mass of @xmath050000@xmath32m@xmath2 .
if we include the source e6 in the above described analysis , the required black hole masses would at least double , due to the increased velocity gradient across the complex .
hence , we conclude that it is safe to assume that the required mass to bind the members of the irs 13e complex must be at least @xmath33m@xmath2 . assuming the extreme case that the 4 brightest stars in irs 13e are similarly massive as the binary irs 16sw ( @xmath34100m@xmath2 , see * ? ? ?
* ) , it appears conservative to assume that the stellar mass in irs 13e does not exceed @xmath35m@xmath2 and can thus be neglected in this analysis .
irs 13e is located at a projected distance of @xmath0@xmath36 or @xmath0130mpc from the @xmath37m@xmath2 black hole sgr a*. it may be associated with the counter - rotating disk of young stars @xcite . in this case , it should be located at @xmath0120mpc behind the plane of the sky . with a radius of @xmath0@xmath38 , a mass of roughly 1000m@xmath2 then would be sufficient to protect the system from tidal disruption . however , as the analysis above shows , the real constraints for binding irs 13e gravitationally result from the intrinsic proper motions of the sources in this complex .
we took into account the unknown location of the hypothetical black hole and various possibilities for the systemic motion of irs 13e in the gravitational potential of sgr a*. in the most conservative case , the minimum mass to bind irs 13e was found to be @xmath39m@xmath2 ( with a significance of 4@xmath28 ) .
this would confine the imbh to a narrow region @xmath40 nw of e1 . outside this region
the required mass is greater than @xmath41m@xmath2 . the lm mass estimator results in even higher masses .
it is generally accepted that accreting black holes are associated with non - thermal radio and x - ray emission .
one may argue that an imbh in irs 13e could be `` starved '' , similar to sgr a * , with its emission below the detection limit of current telescopes .
however , there are several pieces of evidence that the irs 13e cluster is closely associated with the gas and dust of the mini - spiral : first , there is the close spatial relation between irs 13e and emission from warm dust ( see , e.g. , the high - resolution ao 3.8@xmath42 m images in * ? ? ?
present an analysis of the mini - spiral that also supports a close relation between irs 13e and the ism .
also , the stars in irs 13e , with some of them being wr stars , may present favorable sources of gas and dust . and @xcite present spectroscopic evidence for absorption / emission due to gas and dust intrinsic to irs 13e and for interaction between winds from irs 13e and the surrounding ism .
hence , the situation for a hypothetical black hole in irs 13e appears to be different from sgr a * , with plenty of material available for accretion . as for mm / radio emission , @xcite
show that the 13 mm point source observed by @xcite in the irs 13e cluster is most likely due to thermal emission from the 4@xmath42m - excess source e3c in the center of irs 13e , a dusty wr star ( see also * ? ? ?
the two components e3a and e3b are also proposed as dusty wr stars in the analysis by . as for x - ray emission , @xcite report an x - ray source that they associate with irs 13e
. using the position of this source and of sgr a * as listed in their table 1 , we located this x - ray source in the nir reference frame .
its position is marked with a 1@xmath28 error box in fig .
[ fig : propmot ] .
the x - ray source appears offset from irs 13e as also discussed by .
interpret the spectrum of the source as consistent with that of a post - lbv wr colliding wind binary ( although none of the bright irs 13e members appears to be associated with the source , see fig . [
fig : propmot ] ) .
as for a very recent analysis of this x - ray source , it appears to be located a few @xmath43 further north than indicated in fig .
[ fig : propmot ] , so that e5 lies within the positional uncertainty .
no variability of the source was found and it appears to be explained well by colliding winds ( f. baganoff , priv .
. therefore there appears to be no compelling evidence for an accreting imbh in irs 13e .
furthermore , spectroscopy at 4@xmath42 m and 5@xmath42 m @xcite does not reveal any strong broad lines or emission from highly excited ions as they might be expected near an accreting black hole .
infall and dissolution of a massive cluster into the central parsec seems an attractive hypothesis for delivering young stars to the region near sgr a*. generally , the conditions required for a cluster to form near the gc and to fall in towards the central parsec to deposit young , massive stars there within their lifetime , are extreme , i.e. , high core densities and large cluster masses .
an imbh in the core of such a cluster was suggested by @xcite as a means of increasing the efficiency of dynamical friction on the cluster , and thus to relax somewhat the demands on its total mass and core density .
@xcite , however , conclude that an imbh in the cluster core must comprise an unrealistic @xmath010% of the total cluster mass in order for this mechanism to be effective . from the proper motions presented here ,
the required minimum mass for a hypothetical imbh in order to bind the members of irs 13e was derived to be @xmath44m@xmath2 . @xcite
analyze the growth of an imbh by runaway collisions in a dense stellar cluster and conclude that such an object may form in dense clusters and contain of the order @xmath45 of the total stellar mass .
given the high mass estimate for an imbh in irs 13e derived in this work , this would mean an unrealistically high mass of @xmath46m@xmath2 for the progenitor cluster . in this context
it is also important to point out the results of @xcite that exclude the existence of a secondary black hole with masses @xmath44m@xmath2 and semi - major axes between @xmath35 and @xmath47au ( corresponding to @xmath00.1@xmath4 to 12.5@xmath4 angular distance from sgr a * ) .
we can not yet exclude an imbh of up to a few thousand solar masses in irs 13e , but our results suggest that in this case irs 13e must be in the process of dissolution . in this context , we would like to point out the recent work by levin , wu and thommes ( astro - ph/0502143 ) .
they have simulated the infall and dissolution of a stellar cluster with an imbh in the gc and concluded that a hypothetical imbh that binds the irs 13e complex could not have delivered all of the young stars in the gc to their present location .
the positions and proper motions of the irs 13e stars are consistent with their being part of the counter - clockwise disk / ring of young stars ( * ? ? ?
* see also t. paumard et al .
, 2005 , in preparation ) . as for the unusual clustering of the stars in irs 13e , the complex may be , on the one hand , a ( quasi-)permanent feature , i.e. gravitationally bound by a hypothetical imbh .
our analysis shows that in this case , if the star e1 is indeed part of the complex , the imbh would have to be unsually massive .
this makes its existence questionable because of the required large mass of the progenitor cluster and because of the lack of a clear identification of a non - thermal , variable x - ray source centered within irs 13e .
however , we can not completely discard the imbh hypothesis .
on the other hand , irs 13e may be a temporary feature , i.e. , either a chance association ( which is appears somewhat unlikely , even if the stars are part of a disk / ring ) or a cluster in the process of dissolution . at this point
, we would like to point out that an unusual concentration of co - moving stars has recently also been identified among the members of the irs 16 complex in the clockwise rotating disk of stars ( see * ? ? ?
detailed future studies of the proper motions , line - of - sight velocities , distribution , and spectral types of the stars in the central parsec of the gc nuclear cluster are clearly needed in order to understand better the phase - space clustering of the stars in the irs 13e and irs 16 complexes . _ based on observations obtained at the gemini observatory , which is operated by the association of universities for research in astronomy , inc .
, under a cooperative agreement with the nsf on behalf of the gemini partnership : the national science foundation ( united states ) , the particle physics and astronomy research council ( united kingdom ) , the national research council ( canada ) , conicyt ( chile ) , the australian research council ( australia ) , cnpq ( brazil ) and conicet ( argentina ) . _ 3 @xmath28 uncertainty of the direction of its proper motion .
the lengths of the arrows correspond to the magnitude of the respective velocity .
the dashed box marked with an `` x '' designates the position ( 1@xmath28 uncertainty ) of the x - ray source near irs 13e reported by @xcite .
[ fig : propmot],scaledwidth=80.0% ] . shown
are the residual proper motions of the irs 13e sources after subtraction of the average projected velocity ( indicated by the dashed arrows ) of the sources e1 , e2 , e3 , and e4 .
the opening angles define the 1@xmath28 uncertainty of the directions .
the superposed light grey arrows give the magnitude of the velocities after subtraction of a 1@xmath28 error .
right panel : minimum mass required for an imbh in order to gravitationally bind irs 13e .
this mass depends on the location of the imbh and was calculated with the residual velocities of the irs 13e sources .
contours at 7 , 8 , 9 , 10 , 15 , 20 , and 50@xmath48m@xmath2 , growing larger toward lighter shades of gray .
the significance of the masses is 4 - 16@xmath28 .
[ fig : masses],scaledwidth=80.0% ] ccccc source i d & ra [ @xmath4 ] & dec [ @xmath4 ] & v@xmath50 [ kms@xmath51 & v@xmath52 [ kms@xmath51 + e1 & -2.961 & -1.645 & -206@xmath534 & -51@xmath534 + e2 & -3.171 & -1.732 & -311@xmath534 & 77@xmath538 + e3 & -3.191 & -1.527 & -265@xmath5317 & 171@xmath5311 + e4 & -3.210 & -1.411 & -210@xmath539 & 124@xmath5316 + e6 & -3.013 & -1.422 & -128@xmath5316 & 41 @xmath5313 + | the irs 13e complex is an unusual concentration of massive , early - type stars at a projected distance of @xmath00.13pc from the milky way s central supermassive black hole sagittarius a * ( sgr a * ) . because of their similar proper motion and their common nature as massive , young stars it has recently been suggested that irs 13e may be the remnant of a massive stellar cluster containing an intermediate - mass black hole ( imbh ) that binds its members gravitationally in the tidal field of sgr a*. here , we present an analysis of the proper motions in the irs 13e environment that combines the currently best available data with a time line of 10 years .
we find that an imbh in irs 13e must have a minimum mass of @xmath010@xmath1m@xmath2 in order to bind the source complex gravitationally .
this high mass limit in combination with the absence so far of compelling evidence for a non - thermal radio and x - ray source in irs 13e make it appear unlikely that an imbh exists in irs 13e that is sufficiently massive to bind the system gravitationally . |
since the classic works @xcite by frank zerilli in early @xmath1 s on the particle falling in a schwarzschild geometry , a lot of research and study has been performed on this fundamental problem @xcite .
one of the earliest computational calculations was made by press and his co - workers , which is now known as drpp calculation @xcite on the radiation emitted when a particle starting from rest at infinity falls into a non - spinning black hole .
the collision of two black holes is , in principle , one of the most efficient mechanisms for the generation of the gravitational waves . in recent years
the extreme mass ratio limit of the binary system has been a special focus of research in gravitational physics .
extreme - mass - ratio inspirals ( emris ) are one of the main sources of the gravitational waves for the gravitational wave detectors , such as laser interferometer space antenna ( lisa ) @xcite .
emris are binary systems composed of a stellar compact object ( sco ) with a mass , @xmath2 in the range of @xmath3 inspiralling into a massive black hole ( mbh ) with a mass , @xmath4 in the range of @xmath5 located at the galactic center .
thus , the mass ratios involved are @xmath6 . during the slow inspiral phase the system is driven by the emission of gravitational radiation ,
the general features of which are now well understood .
press showed that there is always an intermediate stage where the ringdown is dominated by a set of oscillating and exponentially decaying solutions , quasinormal modes ( qnms ) whose spectrum depends only on the mass of the black hole and the multipole - moment index @xmath7 of the initial perturbation @xcite .
this regime is followed by a power - law _ tail _
decay due to backscattering .
for the emri , the small companion black hole is modeled as a point particle , and the problem can be framed by using the black hole perturbation theory . moreover , as the first approximation , the point particle follows the geodesics in the space - time of the central black hole .
the frequency - domain approach to this problem has achieved many remarkable results .
specifically the accurate @xmath8 determination of the energy flux of gravitational waves was obtained in the frequency - domain @xcite .
however , the frequency - domain approach can take long computational time and lose accuracy for non - periodic orbits ( for example , parabolic orbits , orbits with high eccentricity or decaying orbits ) .
the time - domain approach seems better suited for such orbits @xcite . for the time - domain approach , the finite - difference ( fd )
method is one of the most popular numerical methods .
the fd time - domain methods , however , suffer from the relatively poor accuracy at the moment @xcite unless a very high computational resolution is used .
the main reason is the point particle approximation , i.e. the approximation of the singular source terms .
various approaches to this issue have been attempted , including the regularizing the dirac @xmath0-function using a narrow gaussian distribution and also using more advanced discrete @xmath0-model @xcite .
another approach of the emri problem is to use the spectral ( sp ) method @xcite . in our previous work
, we used the spectral method to solve the inhomogeneous zerilli equation in time - domain and obtained good results .
but the proper power - law decay was not observed @xcite . in early time
the solution agrees with the established solution but in very late time the solution is contaminated by the small - scale oscillations .
these oscillations are likely due to the artificial truncation of the computational domain . in this work
, we continue our previous research with the spectral method in order to obtain the proper power - law decay . for this
, we developed the multi - domain hybrid method .
the multi - domain method hybridizes the spectral method and the high - order finite - difference method . the spectral domain is also split into many sub - domains , each of which is also a spectral domain .
the main advantage of the multi - domain method is that the computational costs can be significantly reduced by reducing the order of the interpolating polynomial in each sub - domain and the parallelization becomes robust .
a fundamental reason for considering the multi - domain method is also to reduce the boundary effects due to the artificial truncation of the computational domain for obtaining the proper late time decay profile of the gravitational waveforms . in order to obtain the proper power - law decay
, the outer boundary needs to be placed afar , in general .
however , having a large size of the computational domain increases the computational costs significantly . in this work ,
we add the finite - difference domain as the boundary domain .
the spectral method is a global method and it is highly sensitive to the boundary effects . to prevent the `` fast '' propagation of these boundary effects , we use a local method instead as the boundary domain , such as the finite - difference domain . by doing this
, we obtain the proper power - law decay while having the computational costs reduced and also exploiting the accuracy of the spectral method . to patch each sub - domain with others ,
we derive the accurate and stable patching conditions . for the spectral and finite - difference sub - domains ,
we show that the resolution across the interface needs to be closely uniform .
otherwise , the cfl condition becomes strict . for the singular source term
, we use both the gaussian @xmath0-function method and the discrete @xmath0-function method . for the gaussian method
, we change the shape of the gaussian profile to mimic the @xmath0-function . for the discrete @xmath0-function
, we generalize the discrete @xmath0-function developed by sundararajan et al .
@xcite into the one on the non - uniform grid .
we provide numerical results that show the efficiency and robustness of the proposed hybrid method . using the hybrid method
we could obtain the proper power - law decay with the gaussian approximation model .
we use various shapes of the gaussian profile and found that the result is insensitive to the shape .
that is , even a broad profile , which results in a smooth solution , yields the power - law decay successfully . with the smooth solution
, the spectral method does not need to use the filter operation , which increases the computational efficiency further .
we also obtain the power - law decay with the discrete @xmath0-function model , but the computed slope was not accurate , which may imply that the discrete @xmath0-function model yields correct results only on uniform grids .
this paper is organized as follows . in section 2 ,
we briefly describe the finite - difference and spectral methods . for the finite - difference method
, we used the @xmath9th - order method . for the spectral method we use the chebyshev spectral collocation method based on the gauss - lobatto collocation points . in section 3
we describe the discrete @xmath0-function on non - uniform grids .
section 4 explains the zerilli equation briefly . in section 5
, we describe the proposed hybrid method in details .
we derive the stable and accurate interface patching conditions .
boundary conditions are described in section 6 . in section 7
, we discuss the stability of the hybrid method . in section 8 ,
numerical results are provided . in section 9 , a brief summary and future work are explained .
in this work we consider both the 2nd and 4th - order finite - difference method . the 2nd - order finite - difference method for the spatial derivatives
are well known and we omit those formulae .
instead we briefly explain the 4th - order finite - difference method . for the 4th - order method we will derive the formula when the grid is non - uniform .
this is because , in the spectral domain , we use the gauss - lobato collocation points , which are not evenly distributed and we need the finite - difference formula for the boundary conditions in the spectral domain .
also we define the modified flux at the sp - sp interface , which also requires the finite - difference formulae .
details of these are described in section @xmath10 .
_ 4th - order finite - difference method : uniform grids .
_ let @xmath11 be the grid spacing in the finite - difference domain and let @xmath12 be @xmath13 .
the standard @xmath14-order derivatives are given by centered difference : @xmath15 off - centered ( 1 - point ) difference : @xmath16 off - centered ( 2 - points ) difference : @xmath17 at the left @xmath18 and right @xmath19 boundaries , we used the 2nd - order difference method .
@xmath18 : @xmath20 @xmath21 similarly @xmath19 : @xmath22 @xmath23 the centered 2nd - order derivative is given by @xmath24 for the spectral method , we use the chebyshev spectral collocation method based on the gauss - lobatto collocation points .
the chebyshev spectral collocation method seeks a solution in the chebyshev polynomial space by the chebyshev polynomials @xmath25 as @xmath26 where @xmath27 is the chebyshev polynomial of degree @xmath28 and @xmath29 the corresponding expansion coefficient .
the commonly used collocation points are the gauss - lobatto quadrature points @xmath30 given by @xmath31 these collocation points belong to @xmath32 $ ] and so we need the proper mapping to fit in our computational domain .
the required mapping is @xmath33 where @xmath34 $ ] is the original computational domain .
the expansion coefficients are given by @xmath35 where @xmath36 if @xmath37 , and @xmath38 otherwise .
we also use the spectral filtering method to minimize the possible non - physical high frequency modes .
the oscillations with the spectral method possibly found near the local jump discontinuity and also generated due to inconsistent initial conditions propagate through the whole domain .
our filtered approximation is given by @xmath39 where @xmath40 based on the exponential filter is the filter function according to @xcite .
our filter matrix @xmath41 is given by @xmath42 where @xmath43 is the order of filteration and @xmath44 is a constant .
the filtered solution at @xmath45 is given by @xmath46
discrete @xmath0-function on a uniform grid has been derived in @xcite . in sp - fd approach
, the singular source term is always located inside the spectral domain . since in the spectral domain
the grid is non - uniform @xcite , we need to redefine the discrete @xmath0-function on the non - uniform grid . @xmath0-function which exists at @xmath47 , @xmath48 , by using the following relation @xmath49 equating the coefficients of @xmath50 from both sides yields @xmath51 for getting the first derivative of the @xmath0 function we have @xmath52 equating the coefficients of @xmath50 from both sides yields @xmath53 % \nonumber % \end{eqnarray } & % \\ \\
\frac{(\alpha - x_{k})(\alpha - x_{k+2})}{(x_{k+3}-x_{k+2})(x_{k+3}-x_{k+1})}[\frac{\alpha - x_{k+1}}{(x_{k+4}-x_{k+2})(x_{k+3}-x_{k})(x_{k+3}-x_{k+2 } ) } & \mbox { at } x_{k+2}\\ \quad\quad -\frac{\alpha - x_{k+3}}{(x_{k+2}-x_{k})(x_{k+1}-x_{k})(x_{k+2}-x_{k+1 } ) } ] & \\ \\ \frac{(\alpha - x_{k})(\alpha - x_{k+1})(\alpha - x_{k+3})}{(x_{k+3}-x_{k+1})(x_{k+2}-x_{k})(x_{k+2}-x_{k+1})(x_{k+3}-x_{k+2})(x_{k+4}-x_{k+3 } ) } & \mbox { at } x_{k+3 } \\ \\ -\frac{(\alpha - x_{k})(\alpha - x_{k+1})(\alpha - x_{k+2})}{(x_{k+5}-x_{k+4})(x_{k+4}-x_{k+2})(x_{k+3}-x_{k})(x_{k+3}-x_{k+1})(x_{k+3}-x_{k+2
} ) } & \mbox { at } x_{k+4}. \end{array } \right.\ ] ] the gaussian @xmath0-function is defined by @xmath54 the shape of the @xmath0-function depends on the full width at half maximum @xmath55 . if @xmath55 is small , @xmath56 is narrow and if @xmath55 has higher value then @xmath56 is broader .
the value of @xmath57 depends on time and determines the position of the @xmath0-function .
the lowest order perturbation theory of the initial schwarzschild black hole spacetime leads to the inhomogeneous zerilli equation with even - parity @xcite .
such an equation describes the gravitational wave @xmath58 in @xmath59 dimension given by the following second - order wave equation @xmath60 where @xmath61 is the tortoise coordinate , @xmath62 the potential term and @xmath63 the source term . for details
see [ 3,10 ] .
the tortoise coordinate , is given by @xmath64 where @xmath65 is the physical coordinate and @xmath66 .
one can convert eq ( 9 ) into a system of equations . in some previous works ,
zerilli equation was solved in time - domain @xcite and in all cases the authors converted the @xmath67nd - order pde to the system of @xmath68st - order equations in space and time .
we understand that some instabilities arose and to suppress those instabilities , they introduced an auxiliary field which converted the equation to a coupled set of @xmath68st - order equations in space and time . in this work
we do not convert the @xmath67nd - order pde to the system of equations .
instead we control the instability efficiently by introducing new interface conditions between sp - sp interfaces and sp - fd interfaces .
we also correct the numerical flux accordingly .
the hybrid method is basically the domain decomposition method .
we use the finite - difference and the spectral methods for the hybrid method .
the one - dimensional single domain is decomposed into multi - domains .
each domain is carried by either the finite - difference method or the spectral method .
let @xmath69 $ ] be the original domain .
suppose that @xmath70 is decomposed into @xmath4 sub - domains and each sub - domain meets its adjacent sub - domain at its boundaries only such that @xmath71 where @xmath72 is the @xmath73th sub - domain and @xmath74 and @xmath75 denote the right and left boundaries of the sub - domain .
let @xmath76 and @xmath77 denote the @xmath73th sub - domain of the sp domain and the fd domain respectively .
the @xmath73th sub - domain @xmath72 should be either @xmath76 or @xmath77 .
let @xmath78 be the total number of grid points of @xmath72 and let @xmath79 be the set of grid points in @xmath72 .
the sub - domains are in order , indexed by the index @xmath73 from @xmath68 to @xmath4 .
that is , @xmath80 is corresponding to the leftmost sub - domain and @xmath81 the rightmost sub - domain . in this work , by default , we let the last domain be the fd domain , i.e. @xmath82 .
the reason is to control the boundary effects arising from the outer boundary .
we found some small oscillations arise from the outer boundary that contaminate the power - law decay when the multi - domain spectral method was used in late times . since the fd method is the local method , we put a fd domain as the boundary domain to remove the boundary effect .
the main development of the hybrid method is the stable patching algorithm at the domain interfaces between the sp and fd domains and the sp - sp domains .
this involves the approximation of the derivatives across the non - uniform grid points , for which we consider the 4th - order fd method with the non - uniform grid and the polynomial interpolation method . the 4th - order centered fd method uses @xmath10 points and the non - uniformity needs to be addressed in the fd approximation of the derivatives
while the 2nd - order centered fd method does not need such non - uniform adaptation .
the schematic diagram is given in diagram 2 . _
2nd - order finite - difference derivative at the spectral domain boundary : _ fd domain imposes the outflow boundary condition @xmath83 at the right domain boundary . in the case that the last domain is a sp domain , we adopt the non - uniform fd method for the term of @xmath84 in the outflow boundary condition .
let @xmath85 and @xmath86 and assume that @xmath87 and @xmath88 are all non - zero constants and are not necessarily the same . with these definitions , let @xmath89 hereafter let _ end _ and _ ghost _ denote the indices of the last grid points and the ghost grid points respectively .
then the derivative @xmath90 at @xmath91 for @xmath92 is given by @xmath93 _ 4th - order finite - difference derivative at the interface across the adjacent spectral domains : _ at the domain interfaces across the sp domains , we use the 4th - order fd method for the 2nd - order derivative ( see diagram 1 ) .
let @xmath94 , @xmath95 and @xmath96 then we have the 2nd - order derivative at @xmath97 @xmath98 using the above relation , the derivative at @xmath99 for the sp domain @xmath76 is given by @xmath100 for the interface patching between the fd domain and the sp domain , we use the high - order interpolation based on the lagrange interpolation .
for the interpolation at the ghost cells in the fd domain for the sp approximation in @xmath76 , we use the @xmath9th - order polynomial .
for example , for fd - sp patching we use the @xmath10 points in the fd domain , @xmath101 .
these points are @xmath102 , @xmath103 .
the points for the ghost cells to seek are @xmath104 and @xmath105 ( two grid points marked with the symbol @xmath106 in diagram 2 ) and these are satisfying the following relations @xmath107 we use the lagrange interpolation to find the interpolation @xmath108 at @xmath109 , and @xmath110 such that @xmath111 for the interpolation at the ghost cells in the sp domain for the fd domain ( @xmath77 , e.g. ) we use the chebyshev interpolation in the spectral domain ( @xmath112 ) .
the grid points for the ghost cells in the sp domain are @xmath113 and @xmath114 and these are satisfying the following relations ( see diagram 2 ) @xmath115 since the fd domain has the uniform grids , we have @xmath116 .
similarly to the fd interpolation , the interpolation @xmath117 at @xmath118 , and @xmath119 is given by for @xmath120 @xmath121 note that @xmath12 is the linear mapping from @xmath122 $ ] to @xmath123 $ ] . here
@xmath124 is the lagrange interpolation polynomial based on the chebyshev polynomials given by @xmath125 where @xmath126 if @xmath127 and @xmath128 otherwise @xcite and @xmath129 is the derivative of the chebyshev polynomial of degree @xmath130 with respect to @xmath131 .
since the polynomial order in each sp domain is low , the interpolation can be done quickly without using the fast transformation . to explain the patching procedure , we consider the following 2nd - order wave equation , @xmath132 . including the potential term is straightforward .
: consider the two adjacent sp domains @xmath112 and @xmath76 whose solutions are denoted by @xmath133 and @xmath58 , respectively .
( hereafter , let @xmath133 and @xmath58 denote the solution in the left and the right sub - domains , respectively . )
the patching is based on eq .
( [ equation20 ] ) at the domain boundaries only . for @xmath112
the left boundary value is updated based on the following @xmath9th - order method : @xmath134 where the superscripts @xmath135 , @xmath136 and @xmath137 indicate the time steps @xmath138 , @xmath139 , and @xmath140 and @xmath141 and @xmath142 .
note that we let every sp domain has the same domain interval .
thus @xmath143 and @xmath144 . 0.5 in ( 1,1 )
( 1.25,1.25)(-1,-1)0.2 ( 0.75,1 ) ( 0.9,1 ) ( 1.1,1 ) ( 1.25,1 ) ( 0.1,1 ) ( 0.5,1 ) ( 0.75,1 ) ( 0.9,1 ) ( 1.0,1 ) ( 1.0,1 ) ( 1.1,1 ) ( 1.25,1 ) ( 1.5,1 ) ( 1.9,1 ) ( 1.27,1.25)@xmath145 ( -0.1,1)(5,0)2.3 ( 0,0.5)@xmath146 ( -0.1,0.3)diagram 1 : sp - sp patching .
grids at the sp and sp interface [ diagram1 ] for @xmath76 the right boundary value is updated in the same way : @xmath147 ( 1,1 ) ( -0.2,1 ) ( 0.2,1 ) ( 0.6,1 ) ( 1.0,1 ) ( 1.0,1 ) ( 1.1,1 ) ( 1.25,1 ) ( 1.5,1 ) ( 1.9,1 ) ( 2.5,1 ) ( 1.25,1.25)(-1,-1)0.2 ( 1.27,1.25)@xmath148 ( 0.9,0.98)@xmath106 ( 0.75,0.98)@xmath106 ( 1.4,0.98)@xmath149 ( 1.8,0.98)@xmath149 ( -0.3,1)(7,0)3 ( 0,0.5)@xmath150 ( -0.1,0.3)diagram 2 : fd - sp patching .
grids at the fd and sp interface .
[ diagram2 ] : for the fd and sp domain patching , we use the polynomial interpolation for the ghost cells . for this case , the @xmath9th - order method is used based on eq . 6 for the uniform grid .
let @xmath58 and @xmath133 be the solutions in the fd domain @xmath77 and sp domain @xmath112 , respectively .
the value at the right boundary of @xmath77 is updated using the following : @xmath151 where @xmath152 and @xmath153 are from eq .
( [ spinterpolation ] ) @xmath154 the value at the left boundary of @xmath112 is updated in the similar way , but using the non - uniform formula eq .
( [ equation25 ] ) , we have @xmath155 where @xmath156 and @xmath157 are defined in the same way as in eq .
( [ equation25 ] ) . and the interpolation values @xmath158 are from eq .
( [ fdinterpolation ] ) given by @xmath159 where @xmath160 and @xmath161 : after the patching procedure for the fd - sp and sp - sp patching , we consider the flux correction .
since we set each sp domain to have the same resolution , it is obvious that @xmath162 @xmath163 we use the central ( average ) flux to get @xmath164\nonumber \\ & = & \frac{1}{2}[2\psi^{j}(end)-\psi^{j-1}(end)+2\phi^{j}(1)-\phi^{j-1}(1 ) ] \nonumber \\ & + & \frac{1}{2}(\frac{\delta t}{h})^{2}[2\phi^{j}(2)+2\psi^{j}(end-1)-2\psi^{j}(end)-2\phi{j}(1 ) ] \nonumber \\ & = & \frac{1}{2}[2\psi^{j}(end)-\psi^{j-1}(end)+2\phi^{j}(1)-\phi^{j-1}(1)]+\nonumber \\ & & \frac{(\delta t)^2}{h}[\frac{\phi^{j}(2)-\phi{j}(1)}{h}-\frac{\psi^{j}(end)-\psi^{j}(end-1)}{h } ] \nonumber \\ & = & \frac{1}{2}[2\psi^{j}(end)-\psi^{j-1}(end)+2\phi^{j}(1)-\phi^{j-1}(1)]+\frac{(\delta t)^2}{h}[\phi_{x}(1)-\psi_{x}(end)].\nonumber\end{aligned}\ ] ] if the solution is smooth , e.g. @xmath165 then @xmath166 .
then the flux reduces to @xmath167.\ ] ] for example , at steady state @xmath168 and so the flux is @xmath169.$ ] similarly for the 4th - order method , @xmath170 \nonumber\end{aligned}\ ] ] and @xmath171\nonumber .
\end{aligned}\ ] ] for sp - sp hybrid process , and at the @xmath136th interface @xmath172 , @xmath173 , @xmath174 and @xmath175 .
we can write the central flux at the @xmath176 interface as follows @xmath177.\end{aligned}\ ] ] if we use the central flux directly then @xmath178 where @xmath179 is the spectral differentiation matrix .
@xmath180+\nonumber \\ & & \frac{(\delta t)^2}{2}[d^2\psi^j_{end } + d^2\phi_{1}^{j}].\end{aligned}\ ] ] if the function is not @xmath181 then @xmath182 . for the fd - sp domain
the resolutions are usually different , i.e @xmath183 and @xmath184 .
but we can expand @xmath185 and @xmath186 around the @xmath187 and @xmath188-th point respectively .
for the boundary conditions at the boundaries @xmath189 of the whole domain @xmath70 , we use the simple outflow condition based on the assumption that the potential term at @xmath190 is negligible .
that is , we use @xmath191 for the first - order derivative in the outflow equations , we use the 2nd - order finite - difference method , eq .
( [ left2ndorder ] ) for the fd domain @xmath192 and eq .
( [ right2ndorder2 ] ) for the sp domain @xmath193 .
0.2 in ( 1,1 ) ( 1.25,1.25)(-1,-1)0.2 ( 0.75,1 ) ( 0.95,1 ) ( 1.1,1 ) ( 1.1,1 ) ( 1.25,1 ) ( 1.45,1 ) ( 1.27,1.25)@xmath194 ( -0.1,1)(5,0)2.3 ( 0,0.5)@xmath195 ( -0.1,0.3)diagram 3 : sp - sp patching .
grids at the sp and sp interface .
[ diagram3 ] for stability analysis we consider sp - sp domain for example .
this analysis is for two domains .
let us consider the two spectral domains denoted by sp1 and sp2 .
let @xmath133 be the solution in sp1 and @xmath58 be the solution in sp2 of the zerrilli equation without the potential and singular source terms .
then we have @xmath196 , \nonumber \\ \psi^{n+1}_0 = 2\psi^{n}_0-\psi^{n-1}_0+\frac{(\delta t)^2}{2h^2}[\phi^{n}_{n-1}-2\psi^{n}_{0}+\psi^{n}_{1 } ] , \nonumber \\ \psi^{n+1}_{j } = 2\psi^{n}_{j}-\psi^{n-1}_{j}+(\delta t)^2\tilde{d_{2}^2}\psi^{n}_{j } , \ : j = 1,\cdots , n-1.\nonumber\end{aligned}\ ] ] we assume that there is no boundary effect . also we consider the equal length intervals and so @xmath197 where @xmath179 is the original chebyshev differential matrix and @xmath198 is the length of the spectral sub - domains .
@xmath199 is the sub - matrix of @xmath200 without the last and first rows and the first column and @xmath201 is the sub - matrix of @xmath202 without last row , first row and first column .
+ collecting all the above equations in matrix form yields , @xmath203 = 2\left[\begin{array}{c } \phi^{n}_{1 } \\
\phi^{n}_{2 } \\ \vdots \\
\phi^{n}_{n}\\\psi^{n}_{0}\\ \vdots \\
\psi^{n}_{n-1 } \end{array}\right ] - \left[\begin{array}{c } \phi^{n-1}_{1 } \\ \phi^{n-1}_{2 } \\ \vdots \\ \phi^{n-1}_{n}\\\psi^{n-1}_{0}\\ \vdots \\ \psi^{n-1}_{n-1 } \end{array}\right]\ ] ] @xmath204 \left[\begin{array}{c } \phi^{n}_{1 } \\ \phi^{n}_{2 }
\\ \vdots \\ \phi^{n}_{n}\\\psi^{n}_{0}\\ \vdots \\
\psi^{n}_{n-1 } \end{array}\right],\ ] ] where @xmath205 is the @xmath206 null matrix .
the previous matrix equation can be written in compact form @xmath207 or @xmath208 eqs @xmath209 and @xmath210 can be written in matrix form @xmath211=\left[\begin{array}{cc}2i_{2n}+\tilde{d } & -i_{2n } \\i_{2n } & 0 \end{array}\right ] \left[\begin{array}{c}w^{n}\\w^{n-1}\end{array}\right].\ ] ] let @xmath212.\ ] ] for the matrix stability , we have the spectral radius of @xmath213 less than @xmath68 .
i.e @xmath214 .
+ + to check the stability we consider two spectral domains @xmath215 $ ] and @xmath216 $ ] .
for these two domains @xmath217 and @xmath218 must be equal .
let @xmath219 be the minimum grid spacing in the left domain and similarly @xmath220 for the right domain .
here we have @xmath221 .
we found the following results for @xmath222(number of grid points in each domain ) and @xmath223(time step ) for which @xmath214 . 0.5 in ( 1,1 )
( 1.25,1.25)(-1,-1)0.2 ( 0.75,1 ) ( 0.95,1 ) ( 1.1,1 ) ( 1.1,1 ) ( 1.25,1 ) ( 1.4,1 ) ( 1.27,1.25)@xmath224 ( 1.15,0.9)@xmath11 ( 1.30,0.9)@xmath11 ( 0.82,0.9)@xmath219 ( 1.0,0.9)@xmath11 ( -0.1,1)(5,0)2.3 ( 0,0.5)@xmath225 ( -0.1,0.3)diagram 4 : sp - fd patching .
grids at the sp and fd interface .
[ diagram4 ] at the sp - fd interface the grid resolutions between the adjacent sub - domains across the domain interface are different .
so the grid distribution is non - uniform .
the stable interface conditions derived for the uniform grid system are not enough and we need some conditions with which the spatial non - uniformity can be addressed properly [ see 26 and references therein ] .
let @xmath226 be the number of grid points in each sp sub - domain and @xmath227 be the number of grid points in the fd domain .
also assume @xmath228 be the length of each sp sub - domain and @xmath229 be the length of the fd sub - domain .
then for stability we must have @xcite @xmath230 consider one spectral domain ( sp1 ) and one fd - domain ( fd1 ) .
let @xmath133 be the solution in sp1 and @xmath58 be the solution in fd1 of the simple wave equation without any potential term .
let number of grid points in sp1 and fd1 be @xmath226 and @xmath227 respectively which satisfy eqn ( 25 ) .
we assume that there is no boundary effect .
then from spectral collocation and @xmath67nd order fd - method , we have @xmath231 , \nonumber \\ \psi^{n+1}_0 = 2\psi^{n}_0-\psi^{n-1}_0+\frac{(\delta t)^2}{2h^2}[\phi^{n}_{n_{1}-1}-2\psi^{n}_{0}+\psi^{n}_{1 } ] , \nonumber \\ \psi^{n+1}_{j } = 2\psi^{n}_{j}-\psi^{n-1}_{j}+(\delta t)^2\tilde{d_{2}}\psi^{n}_{j } , \ : j = 1,\cdots , n_{2}-1.\nonumber\end{aligned}\ ] ] where @xmath232 where @xmath179 is the original chebyshev matrix , @xmath199 is the sub - matrix of @xmath200 without the first and last rows and the last column and @xmath218 is the @xmath67nd - order fd - differentiation matrix , which is written by @xmath233.\ ] ] + and @xmath234 is the sub - matrix of @xmath218 without the first and last rows and the first column . collecting all the above equations in matrix form yields , @xmath235 = 2\left[\begin{array}{c } \phi^{n}_{1 } \\
\phi^{n}_{2 } \\ \vdots \\
\phi^{n}_{n_{1}}\\\psi^{n}_{0}\\ \vdots \\
\psi^{n}_{n_{2}-1 } \end{array}\right ] - \left[\begin{array}{c } \phi^{n-1}_{1 } \\ \phi^{n-1}_{2 } \\ \vdots \\
\phi^{n-1}_{n_{1}}\\\psi^{n-1}_{0}\\ \vdots \\ \psi^{n-1}_{n_{2}-1 } \end{array}\right]\ ] ] @xmath236 \left[\begin{array}{c } \phi^{n}_{1 } \\ \phi^{n}_{2 } \\ \vdots \\ \phi^{n}_{n_{1}}\\\psi^{n}_{0}\\ \vdots \\
\psi^{n}_{n_{2}-1 } \end{array}\right],\ ] ] where @xmath237 is the @xmath238 null matrix , and @xmath239 is the @xmath240 null matrix .
the previous matrix equation can be written in the following compact form @xmath207 or @xmath241 eqs @xmath242 and @xmath243 can be written in matrix form @xmath211=\left[\begin{array}{cc}2\tilde{i}+\tilde{d } & -\tilde{i } \\\tilde{i } & 0 \end{array}\right ] \left[\begin{array}{c}w^{n}\\w^{n-1}\end{array}\right].\ ] ] let @xmath244.\ ] ] where @xmath245 is the unit matrix of order @xmath246 .
for the matrix stability , we must have the spectral radius of @xmath247 less than @xmath68 .
i.e @xmath248 .
+ + to check the stability we consider the sp - domain @xmath123 $ ] and fd - domain @xmath249 $ ] .
we choose @xmath250 and @xmath251 in such a way that they must satisfy eq .
we found the following results for @xmath250(number of grid points in sp - domain ) , @xmath251 ( number of grid points in fd - domain ) and @xmath223(time step ) for which @xmath248 .
some examples are given in the following table : the table implies that the more uniformity of the grid spacing across the sp and fd interface is achieved the larger value of time step could be used for stability . for our numerical experiments in the next section ,
we choose the geometric parameters for the sub - domains so that the grid resolution is uniform in the small neighborhood across the sp - fd interface . here
note that for the numerical experiments in this paper , we choose the geometric setting so that we have about @xmath252 .
for the numerical experiments , the computational domain is split into two main sub - domains , i.e. the sp and fd domains .
the @xmath0-function is located in the sp sub - domain for all time ( see diagram 5 ) .
the sp sub - domain is also divided into multiple smaller sp sub - domains and each sub - domain communicates with its adjacent sub - domains by the interface condition described in section 5 .
the last sp sub - domain communicates with the boundary fd sub - domain . 1.2 in ( 1,1 )
( 0.15,1.22)(-1,-1)0.2 ( 0.15,1.25 ) sp sub - domains ( 0.80,1.22)(1,-1)0.2 ( 1.38,1.22)(-1,-1)0.2 ( 1.30,1.25 ) fd sub - domain ( 1.90,1.22)(1,-1)0.2 ( 0.70,1)(0,1)0.04 ( 0.9,1)(0,1)0.04 ( 1.1,1)(0,1)0.04 ( 0.5,1)(0,1)0.04 ( 0.3,1)(0,1)0.04 ( 0.1,1)(0,1)0.04 ( -0.1,1)(0,1)0.04 ( 2.2,1)(0,1)0.04 ( -0.1,1)(5,0)2.3 ( -0.1,0.6)diagram 5 : sp - sp .... - sp - fd subdomains . [ diagram5 ] since the spectral method is a global method , the boundary effects spread instantly throughout the domain .
thus , the sp method suffers from the boundary effects unless the range of the domain is large enough .
however , the multiple sp sub - domains with the fd boundary domain help avoid the unphysical oscillations and reduce the computational cost . to determine the number of sub - domains and the order of the interpolating polynomial in each sub - domain ,
two aspects are considered .
the first aspect is the resolution for the singular source term and the second one is the grid uniformity across the sp and fd domain interface . as explained in the previous section , the non - uniformity of the grid resolution near the sp and fd interface makes the cfl condition strict .
considering these aspects , we truncate the domain to @xmath253 $ ] in @xmath254 .
the sp sub - domain is @xmath255 $ ] and the fd sub - domain is @xmath256 $ ] .
the multi - domain sp method reduces the computational time significantly by reducing the size of the system but still achieves the desired accuracy .
the typical setup in this work is that the number of sub - domains in the sp domain is @xmath257 and each domain has the interpolating order of @xmath258 .
the time - step is relatively large such as @xmath259 and the gaussian @xmath0-function has @xmath260 .
the waveform is collected at various values of @xmath61 . .
@xmath261 , @xmath262 , @xmath263.,scaledwidth=110.0% ] first we consider the case with the gaussian model .
the first result is in figure 1 .
the waveform is collected at @xmath264 .
the left figure in the top panel shows the ringdown profile which starts around the time @xmath265 .
the right figure in the top panel shows the power - law decay .
the two figures in the lower panel show the two distinct phases . for the time approximately
@xmath266 the solution decays exponentially and it is oscillatory .
this phase is the quasi - normal - mode ( qnm ) ringing phase . after this phase the power - law decay starts .
according to the seminal work by richard price @xcite , the observer measures the late time perturbation field to drop off as an inverse power law of time , specifically as @xmath267 . in the case of schwarzschild black hole , @xmath268 , where @xmath7 is the multipole moment of the perturbation field . in the context of our present work @xcite , @xmath269 .
the right figure in lower panel shows the power - law decay in logarithmic scale both in the horizontal and vertical axes . in this phase ,
the slope of the decay profile is about @xmath270 .
the slopes of the power - law decay for various @xmath55 and different resolutions , are almost same and @xmath271 .
this agrees well with the theoretical result @xcite : @xmath272 . .
@xmath261 , @xmath262 , @xmath263 .
, scaledwidth=110.0% ] in figure 2 , we use the gaussian @xmath0-function with @xmath273 .
the waveform is collected at @xmath274 .
the top panel shows the ringdown profile and lower panel shows the qnm and the power - law decay .
the similar qnm and the power - law decay are observed .
also we find that the same number of oscillations are observed in the qnm regime as in figure 1 . .
@xmath261 , @xmath262 , @xmath263.,scaledwidth=110.0% ] in figure 3 , we use the gaussian model with @xmath275 .
the waveform is collected at @xmath276 .
note that the higher value of @xmath55 is used for this figure and the gaussian @xmath0-function is much smoother than the previous two cases .
we , however , found that the similar solution with the expected qnm profile and power - law decay profile was obtained . .
@xmath261 , @xmath262 , @xmath263 .
, scaledwidth=110.0% ] in figure 4 , we use the gaussian @xmath0 model with @xmath277 .
the waveform was collected at @xmath278 . with this high value of @xmath55 ,
the gaussian @xmath0-function is even smoother .
this high level of smoothness of the singular source term made it unnecessary to use the spectral filter for the solution .
for the previous figures , we used the spectral filter to regularize the solution due to the possible gibbs oscillations . without the filter operation , the computational time
was reduced .
this suggests that the desired decaying profile can be obtained by having the parameters in a clever way .
for example , in figure 5 , we show the polynomial order @xmath279 that we used for figures with different values of @xmath55 to obtain the desired qnm and the power - law decay .
the parameter values in the figure are @xmath280 , @xmath281 , @xmath282 , @xmath283 . in our current research
we did nt attempt , but it would be an interesting study to investigate the optimal configuration of these parameters . in each sp sub - domain with the value of @xmath55 for the gaussian model .
@xmath284.,scaledwidth=60.0% ] we repeated the above experiment with the discrete @xmath0-function . by definition , the discrete @xmath0-function is localized and its shape changes with time because the spectral grid spacing is non - uniform
. -function .
@xmath261 , @xmath285 , @xmath286.,scaledwidth=110.0% ] - function .
@xmath287 , @xmath288 , @xmath289.,scaledwidth=110.0% ] the qnm and the power - law decay profiles are depicted in fig .
but the slope of the power - law tail does not match with that for the gaussian model .
for example , in fig .
6 we considered the case that @xmath290 , @xmath291 and the interface between the sp and fd domains is at @xmath292 .
we use @xmath293 sp sub - domains with @xmath294 .
the waveform was collected at @xmath295 .
the decay rate is slower than expected .
the estimated slope is @xmath296 while the slope with the gaussian model was close to @xmath297 .
although we increase the resolution , @xmath279 , no significant improvement was observed .
7 shows the qnm and the power - law tail with the discrete @xmath0-function .
here we used the very far outer boundary @xmath298 .
the slope is still about @xmath299 .
it seems that the discrete @xmath0-function is good for the fd domain with the uniform grid but it is not ideal for the non - uniform grid with the sp domain .
in this work , the inhomogeneous zerilli equation was solved numerically in time - domain , for which we developed a multi - domain hybrid method based on the chebyshev spectral collocation method and the @xmath9th - order explicit finite - difference method . using the developed method ,
we computed the waveforms for head on collisions of black holes in one dimension . the in - falling black hole is modeled as a point source and consequently singular source terms appear in the governing equation . for the approximation of singular source terms , we used both the gaussian @xmath0-function and the discrete @xmath0-function . for the stable and accurate approximation
, we derived the interface conditions between the spectral and spectral domains and the spectral and finite - difference domains .
the main approach introduced in this work is the use of the finite - difference domain as the boundary domain . without the finite - difference domain as the boundary domain ,
the multi - domain composed of only the spectral sub - domains does not yield the proper power - law decay profile unless the range of the computational domain is very large . using the multi - domain approach with the finite - difference boundary domain method ,
we could obtain the proper power - law decay profile with a relatively small computational cost .
that is , the cfl condition is much relaxed and the location of the outer boundary of the computational domain is not afar .
numerical results show that the hybrid method obtains a proper quasi - normal and power - law decay with the gaussian @xmath0-function approximation .
interestingly , even with a large value of @xmath55 , the proper power - law decay was observed . with the large value of @xmath55 ,
the spectral filtering operation was not necessary , so the computational time was much reduced . the hybrid method with the discrete @xmath0-function approximation ,
however , did not yield the proper power - law decay .
the current study only considered the multi - domain spectral and finite - difference method with the @xmath0-function residing in the spectral domain .
we will investigate the optimal configuration of the multi - domain computational domain in our future research .
99 f. j. zerilli , effective potential for even - parity regge - wheeler gravitational perturbation equations , _ phys .
_ 24 ( 1970 ) 737 - 738
. f. j. zerilli , gravitational field of a particle falling in a schwarzschild geometry , _ phys .
2(1970 ) 2141 - 2160
. j. l. barton , d. j. lazar , d. j. kennefick , g. khanna , l. m. burko , computational efficiency of frequency and time domain calculations of extreme mass ratio binaries : equatorial orbits , _ phys .
d _ 78 ( 2008 ) 064042 .
p. a. sundararajan , g. khanna , s. a. hughes , towards adiabatic waveforms for inspiral into kerr black holes : a new model of the source for the time domain perturbation equation , _ phys .
d _ 76 ( 2007 ) , 104005(1)-104005(20 ) .
p. a. sundararajan , g. khanna , s. a. hughes and s. drasco , towards adiabatic waveforms for inspiral into kerr black holes : ii .
dynamical sources and generic orbits , _ phys .
d _ 78 ( 2008 ) , 024022(1)-024022(13 ) . | a hybrid method is developed based on the spectral and finite - difference methods for solving the inhomogeneous zerilli equation in time - domain .
the developed hybrid method decomposes the domain into the spectral and finite - difference domains .
the singular source term is located in the spectral domain while the solution in the region without the singular term is approximated by the higher - order finite - difference method .
the spectral domain is also split into multi - domains and the finite - difference domain is placed as the boundary domain . due to the global nature of the spectral method , a multi - domain method composed of the spectral domains
only does not yield the proper power - law decay unless the range of the computational domain is large .
the finite - difference domain helps reduce boundary effects due to the truncation of the computational domain .
the multi - domain approach with the finite - difference boundary domain method reduces the computational costs significantly and also yields the proper power - law decay .
stable and accurate interface conditions between the finite - difference and spectral domains and the spectral and spectral domains are derived . for the singular source term
, we use both the gaussian model with various values of full width at half maximum and a localized discrete @xmath0-function . the discrete @xmath0-function was generalized to adopt the gauss - lobatto collocation points of the spectral domain .
the gravitational waveforms are measured .
numerical results show that the developed hybrid method accurately yields the quasi - normal modes and the power - law decay profile .
the numerical results also show that the power - law decay profile is less sensitive to the shape of the regularized @xmath0-function for the gaussian model than expected .
the gaussian model also yields better results than the localized discrete @xmath0-function . |
the main motivation of our modeling work is to contribute to building realistic prominence models . for this
, we need an accurate knowledge of thermodynamic quantities ( temperature , densities , ) , level populations ( useful , e.g. , to infer the magnetic field properties from spectro - polarimetric observations ) , velocity fields , energy budget .
however these quantities still have large uncertainties associated with them .
observations of several different lines from different atoms / ions allow us in theory to measure these parameters . among these lines ,
the h and he lines are important as they are strong and largely contribute to the radiative losses .
however the prominence plasma being out of lte and optically thick in h and he resonance lines , the interpretation of line spectra or intensities in radially moving prominences is a non - trivial task .
therefore , non - lte radiative transfer calculations including velocity fields are needed to build realistic prominence models . here
we present such calculations and preliminary results .
the prominence is represented by a 1d plane - parallel slab standing vertically above the solar surface .
each prominence model is defined by a set of free parameters : the temperature @xmath0 , the gas pressure @xmath1 , the slab thickness @xmath2 ( or the total column mass ) , the height of the slab above the limb @xmath3 , the microturbulent velocity , and the radial speed . for this preliminary study
we consider isothermal and isobaric prominences , although the code allows for inclusion of a transition region between the cold prominence and the hot corona .
we first solve the pressure equilibrium , the ionization equilibrium , and the coupled statistical equilibrium ( se ) and radiative transfer ( rt ) equations for a 20 levels h atom .
then the se and rt equations are solved for other elements : ( 29 levels ) and ( 4 levels ) , and ( 5 levels ) .
more details on the modeling of the hydrogen , calcium , and helium spectra in quiescent prominences can be found in @xcite respectively , and references therein . for the modeling of active and eruptive prominences
, we use a velocity - dependent incident radiation as boundary conditions for the rt equations .
it has already been shown by @xcite in the case of the hydrogen lines that the doppler effect induces a frequency shift of the incident profile relative to the rest case , and also a distortion of the incident profile due to the variation of the doppler shift with the direction of the incident radiation .
it is also the case for the helium ( @xcite ) and calcium incident radiation .
we reproduce the results of @xcite who computed the hydrogen radiation emitted by a radially moving prominence , using partial redistribution in frequency ( prd ) for the lyman lines .
we obtain the same variation of the relative intensities ( intensities normalised to the line intensities when the prominence is at rest ) and the same line profiles for ly@xmath4 , ly@xmath5 , and h@xmath4 .
the main result is that there exists an important coupling between ly@xmath5 and h@xmath4 which causes these lines to be first doppler brightened , and then doppler dimmed , with increasing velocity , while there is only a doppler dimming effect on ly@xmath4 . figure [ nl - fig : intensities ] presents relative intensities as a function of velocity for the 584 , 304 , and 10830 lines ( left panel ) and k and 8542 lines ( right panel ) at two different temperatures ( 8000k and 15000k ) .
the 10830 line does not show any sensitivity to the doppler effect , which is mainly due to the very weak incident absorption line .
the 584 line is quite sensitive to the doppler effect .
its doppler dimming is more important at low temperature .
the resonance lines are the most sensitive to the radial velocity of the plasma ( the relative intensity of the 256 line , not shown , exhibits a similar variation as 304 ) , and the doppler dimming is strong at the temperatures considered in this study .
such a result was expected since the main mechanism of formation at these temperatures for these lines is the scattering of the incident radiation ( @xcite ) .
let us stress that in this preliminary study we have not included a transition region between the cold prominence and the hot corona ( pctr ) .
the presence of a hotter plasma in the pctr may somehow decrease the sensitivity of the resonance lines to the doppler effect as collisions will become more important in the formation processes of these lines .
this will be investigated in a future work .
the right panel of fig .
[ nl - fig : intensities ] indicates that there is no strong doppler effect on the resonance lines , while we observe some doppler brightening of the 8542 line ( and indeed of the other two infrared lines at 8498 and 8662 , not shown ) at low temperature .
if spectroscopic observations of erupting prominences are available , then a comparison between computed and observed line profiles can be made .
we show in figs .
[ nl - fig : profils he ] and [ nl - fig : profils ca ] the line profiles for the same helium and calcium lines considered in fig . [
nl - fig : intensities ] at two different temperatures ( solid line : 8000 k , dashed line : 15000 k ) , at four different velocities ( from top to bottom : 0 , 80 , 200 , and 400kms@xmath6 ) .
the doppler dimming effect is well observed in the helium resonance lines at 584 and 304 as the radial velocity is increased ( fig .
[ nl - fig : profils he ] ) .
we can observe asymmetries in the line profiles of these lines when the prominence plasma is moving radially , with some intensity enhancement which is especially visible in the red wing of the 584 line at low temperature .
this is explained as follows .
the radiation emitted by the disk center in our code is represented by symmetrical line profiles .
when the prominence is at rest , the incident radiation illuminating the prominence slab is symmetrical ; however when the prominence plasma is moving radially the incident line profile becomes distorted and shifted towards the red . as we used the prd standard approximation in our calculations of the resonance lines of helium ,
the resulting line profiles are asymmetrical .
this would not have been the case if we had considered complete redistribution in frequency ( crd ) instead of prd .
the line asymmetry is more visible for the 584 line as its wings are fairly bright , while the wing intensities of the 304line are low . despite the fact that the line asymmetry increases with speed for both lines , it is more visible at low speeds ( when the intensity in the wing is high enough ) . finally , it is more pronounced at low temperatures because the scattering of the incident radiation is relatively more important as compared to collisional processes than it is at higher temperatures .
figure [ nl - fig : profils ca ] shows that the intensities of the lines are lower at 15000k than at 8000k , an effect of to ionization ( @xcite ) .
k ( solid line ) and @xmath7k ( dashed line ) , with @xmath8 dyn@xmath9 , and @xmath10 km , at different velocities : 0 , 80 , 200 , and 400kms@xmath6 from top to bottom .
abscissa is @xmath11 in and vertical axis is specific intensity in ergs@xmath6@xmath9sr@xmath6@xmath6 . from left to right : 584 , 304 , and 10830.,scaledwidth=69.0% ] for k and 8542 lines.,scaledwidth=69.0% ]
we show here how the non - lte radiative transfer calculations can help us to infer the thermodynamic properties of a prominence observed by the sumer spectrometer on soho .
this prominence was actually a rather quiet prominence and we have not included any velocity fields in these calculations .
it was observed during the 13th medoc campaign held at ias in june 2004 .
we select a few pixels in the sumer slit which cut across the prominence and average the line profiles there .
we consider the line profiles of two h resonance lines ( ly@xmath5 and ly@xmath12 ) and the resonance line at 584 . for the comparison between computed and observed line profiles
we now include the presence of a pctr .
the temperature variation between the cold prominence core and the corona suggested by @xcite has been adopted for this study . by a trial and error process we selected the temperature profile shown in fig .
[ nl - fig : temp ] .
the other prominence parameters are @xmath13 dyn@xmath9 , @xmath14 km ( total column mass 2.4@xmath15g@xmath9 ) , @xmath16 km , and the microturbulent velocity @xmath17 kms@xmath6 .
we obtain a very good agreement between the computed profiles ( convoluted with the sumer instrumental profile ) and the observed profiles , as shown in fig .
[ nl - fig : match ] .
it is worth noting that fitting hydrogen _ and _ helium resonance lines simultaneously places strong constraints on the parameter space , and it was not possible to find another set of parameters for the prominence that would be significantly different than what is given above and that would lead to a satisfactory fit of the observed profiles .
( left ) , ly@xmath12 ( middle ) , and 584 ( right).,scaledwidth=69.0% ]
the non - lte radiative transfer modeling that we are developing is a key tool for interpreting observations and constructing realistic prominence models . the combination of lines from hydrogen , helium , and calcium , places strong constraints on the models .
imaging and spectroscopy must be used for comparisons with calculations to determine thermodynamic parameters and velocities . the radial velocity determined from the comparison between observed and computed line profiles , in combination with line - of - sight velocities , should allow us to infer the full velocity vector of the prominence plasma . in a future work we will compare our model results with simultaneous observations of , e.g. , h@xmath4 and 304 . | active prominences exhibit plasma motions , resulting in difficulties with the interpretation of spectroscopic observations .
these solar features being strongly influenced by the radiation coming from the solar disk , doppler dimming or brightening effects may arise , depending on which lines are observed and on the velocity of the plasma . interlocking between the different atomic energy levels and non local thermodynamic equilibrium lead to non - trivial spectral line profiles , and this calls for complex numerical modeling of the radiative transfer in order to understand the observations .
we present such a tool , which solves the radiative transfer and statistical equilibrium for h , , , and in moving prominences where radial plasma motions are taking place .
it is found that for isothermal , isobaric prominence models , the resonance lines are very sensitive to the doppler effect and thus show a strong doppler dimming .
the lines doppler effect for the prominence models considered here .
we illustrate how the code makes it possible to retrieve the plasma thermodynamic parameters by comparing computed and observed line profiles of hydrogen and helium resonance lines in a quiescent prominence .
this new non - lte radiative transfer code including velocities allows us to better understand the formation of several lines of importance in prominences , and in conjunction with observations , infer the prominence plasma thermodynamic properties and full velocity vector . |
one of the active areas of high energy physics research has been exploration of the so called quantum chromodynamics ( qcd ) phase diagram . in particular , with the advent of ultrarelativistic heavy ion beams at rhic and cern and with the upcoming facilities of gsi where compressed baryonic matter is expected to be produced , such studies have assumed special importance . beside the laboratory experiments , various astrophysical objects like neutron stars , quark stars ,
provide natural sites where many of the theoretical conjectures about the various phases of quark matter can be tested .
the latter , in the present context , is more relevant here , as we study the possibility of para - ferro phase transition in dense quark system interacting via one gluon exchange .
the original idea about para - ferro phase transition in quark matter was proposed recently in @xcite where the possibility of bloch like phase transition @xcite was studied and it was shown that spin polarized quark matter might exist at low density @xcite .
the underlying mechanism of such a phase transition is analogous to what was originally proposed for the degenerate electron gas @xcite .
there , for coulomb interaction , it was shown that the exchange correction to the energy is attractive which at the low density wins over the kinetic energy giving rise to a ferromagnetic state @xcite . in @xcite ,
a variational calculation has been performed to show that it is indeed possible to have a spin polarized quark matter at low density of strange quark system , while for light quark it never happens @xcite .
similar difference of the light and strange quark matter , albeit in a different context , was observed earlier @xcite .
however , in @xcite , it was shown that both the light and heavy flavor systems can exhibit such phase transitions although the critical density for the strange matter is higher than the light quark systems .
such investigations , have also been performed in @xcite and also in @xcite where the calculation has been extended to include thermal effects .
the bloch like phase transition , for strange quark , has also been reconfirmed in @xcite .
one shortcoming of all these works including @xcite , has been that the calculations were restricted to the hartree fock level and the terms beyond the exchange diagrams , commonly termed as correlation energy @xcite were ignored . without such corrections , however , the calculations are known to remain incomplete as the higher order terms are plagued with infrared divergences arising out of the exchange of massless gluons , indicating the failure of the naive perturbation series .
we know that this problem can be cured by reorganizing the perturbation theory where a particular class of diagrams , _ viz . _
the bubbles are resummed in order to obtain a finite result .
originally , as is well known , this was done by gell - mann and brueckner @xcite while calculating the ground state energy of degenerate electron gas .
the contribution of the bubbles involve terms of @xmath5 indicating non - perturbative nature of the correction @xcite . in the present work ,
as announced , we calculate the spin susceptibility @xmath6 of dense quark system with corrections due to correlations _
i.e. _ containing terms upto @xmath5 .
this requires the knowledge of the ground state energy ( gse ) of spin polarized matter with inclusion of bubble diagrams .
the gse of the polarized quark matter has been calculated only recently in @xcite which is the starting point of the present paper .
this work is very similar to that of brueckner and swada @xcite and those of @xcite , applied to the case of qcd matter .
unlike , degenerate electron gas , however , we have both the electric and magnetic interactions and the calculation is performed relativistically , while the non - relativistic results appear as a limit .
the spin susceptibility @xmath0 , for quark matter upto @xmath7 has already been calculated in ref.@xcite which we only briefly discuss .
subsequently , the non - fermi liquid corrections to @xmath0 has also been studied in @xcite .
these studies provide further motivation to undertake the present endeavor to include correlation corrections , without which , as mentioned already , the perturbative evaluation of @xmath8 remains incomplete .
in addition , we also calculate incompressibility and sound velocity for spin polarized quark matter with corrections due to correlations which involve evaluation of single particle energy at the fermi surface .
these quantities are of special interests for applications to astrophysics .
moreover , we also evaluate the exchange energy density at non - zero temperature and determine the corresponding corrections to the spin susceptibility .
the plan of the paper is as follows . in sec .
ii we calculate spin susceptibility with correlation correction for degenerate quark matter .
analytic expressions are presented both in ultra - relativistic ( ur ) and non - relativistic ( nr ) limit . in sec .
iii , we evaluate exchange energy density and spin - susceptibility at non - zero temperature . in sec .
iv we summarize and conclude .
detailed expressions of the intermediate expressions , from which @xmath0 is derived , have been relegated to the appendix .
the spin susceptibility of quark matter is determined by the change in energy of the system as its spins are polarized @xcite .
we introduce a polarization parameter @xmath9 with the condition @xmath10 , where @xmath11 and @xmath12 correspond to densities of spin - up and spin - down quarks respectively , and @xmath13 denotes total quark density .
the fermi momenta in the spin - polarized quark matter then are @xmath14 and @xmath15 , where @xmath16 , is the fermi momentum of the unpolarized matter @xmath17 . in the small @xmath18 limit , the ground state energy behaves like @xcite [ xi_expan ] e()&= & e(=0)+_s^2+o(^4 ) . here ,
@xmath19 , defined to be the spin stiffness constant in analogy with @xcite .
the spin susceptibility @xmath0 is proportional to the inverse of the spin stiffness , mathematically @xmath20 @xcite .
it is to be noted that in eq.([xi_expan ] ) , the first term corresponds to unpolarized matter energy .
now , the leading contributions to the ground state energy are given by the three terms _ viz .
_ kinetic , exchange and correlation energy density @xcite i.e. e & = & e_kin+e_ex+e_corr . the total kinetic energy density for spin - up and spin - down quark becomes @xcite [ rel_kin ] e_kin&= & \{p_f(1+)^1/3 .
+ & & . -m_q^4
( ) + [ - ] } , where @xmath21 is the quark mass . the exchange energy density @xmath22 have been calculated in ref.@xcite within fermi liquid theory approach .
one can also directly evaluate the two loop diagram @xcite to obtain e_ex^nf&=&_s= ( p_f^s-|p|)(p_f^s-|p|)f_pp^nf , [ ex_nf ] + e_ex^f&= & 9 ( p_f^+-|p|)(p_f^--|p|)f_pp^f , [ ex_f ] where @xmath23 and @xmath24 stands for non - flip @xmath25 and flip @xmath26 forward scattering amplitude given in @xcite . here
, @xmath27 can be estimated numerically . however , analytical evaluation of these integrals is possible in the ultra - relativistic and non - relativistic limits as reported in @xcite .
the next higher order correction to the ground state energy beyond the exchange term is the correlation energy @xmath28 @xcite .
the detailed calculation of correlation energy for spin polarized matter have been derived in @xcite which we quote here : [ cor_eng2 ] e_corr & & _ 0^/2 ^ 2_e d_e \{_l^2 + 2_t^2 } , with @xmath29 .
the relevant @xmath30 and @xmath31 are determined to be @xcite _ l & = & _ s= , [ pil_polar ] + _ t & = & _ s = p_f^s^2_e .
[ pit_polar ] the spin susceptibility is given by@xcite ^-1&=&_=0 .
we have @xmath32 .
the kinetic and exchange contribution have been evaluated in ref.@xcite , is given by _
kin^-1&=&. [ chi_kin ] + _ ex^-1&=&- \{2- -+}. [ chi_ex ] + to determine the correlation correction to spin susceptibility , we expand curly braces terms of eq.([cor_eng2 ] ) in powers of the polarization parameter @xmath18 , which gives [ cor_expan ] _ l^2 + 2_t^2&=&(a_0l+b_0t)+^2(a_1l+b_1 t ) + o(^4 )
. + here , @xmath33 and @xmath34 correspond to unpolarized matter term and the detailed expressions of @xmath35 and @xmath36 are given in the appendix .
@xmath37 is [ chi_corr ] _ corr^-1 & = & _ = 0 + & & _ 0^/2 ^ 2_e d_e ( a_1l+b_1 t ) . from the above expression and with the help of the expression presented in the appendix ,
@xmath37 can be estimated numerically .
results for the two limiting cases however can be obtained analytically as we present in the following two sub - sections . in the ultra - relativistic limit ,
the kinetic , exchange and correlation energies are @xcite e_kin^ur&= & , + e_ex^ur&=&p_f^4 , + e_corr^ur & = & p_f^4 [ ( 1+)^4/3+(1-)^4/3 + 2(1-^2)^2/3 ] .
[ ur_eng ] with the help of eq.([xi_expan ] ) , each energy contribution to the susceptibility is [ ur_contri ] _ kin^-1 & = & + _ ex^-1 & = & - + _ corr^-1 & = & -(r_s-0.286 ) . with @xmath38 . from eq.([ur_contri ] ) , sum of all the contribution to the susceptibility can be written as @xcite [ ur_suscep ] ^ur & = & _ p[1 - -(r_s-0.286)]^-1 , where @xmath39 is the non - interacting susceptibility @xcite .
now , we go to the non - relativistic limit to calculate spin - susceptibility in order to compare our results with those of dense electron gas @xcite interacting via .
static coulomb potential . in this limit ,
kinetic and exchange energy densities are @xcite : e_kin^nr&= & , [ kin_nr ] + e_ex^nr&=&-p_f^4 .
[ ex_nr ] the contribution to the susceptibility from kinetic and exchange energy density yields _ kin^-1&= & + _ ex^-1&=&- .
we want to calculate the contribution to the spin - susceptibility beyond the exchange correction . for this
we first evaluate the correlation energy in this limit . the dominant contribution to the correlation energy
is found to be , [ nr_corr ] e_corr^nr&=&- _ ^1/2^k_c _ 0^xdx_s = f(s ) _ s= ( 1-x^s ) , where @xmath40 , @xmath41 , @xmath42 , @xmath43 and @xmath44 . for @xmath45
one obtains : [ corr_nf ] e_corr^nr , s = s & & m_q p_f^3(1 - 2 ) , note that , here , the correlation energy is independent of spin - polarization @xmath18 .
it is seen , that for the spin parallel interactions @xmath18-dependent terms contribute with opposite sign and cancels each other . for @xmath46 ,
the integral on @xmath47 takes the form [ xi2_term1 ] i&=&_0^xdx\{(1 + ) + ( -)}. + expanding @xmath48 in terms of @xmath18 and retain upto @xmath49 we have [ xi2_term2 ] i & & . using eq.([nr_corr ] ) , ( [ xi2_term1 ] ) and ( [ xi2_term2 ] ) we have [ corr_f ] e_corr^nr , s =- s & & m_q p_f^3 . it is to be mentioned that similar expressions for degenerate electron gas interacting via .
static coulomb potential can be found in ref.@xcite . from eq.([corr_nf ] ) and ( [ corr_f ] ) , it is clear that spin anti - parallel states are attractive in contrast to the parallel states due to pauli exclusion principle . in this limit
the correlation contribution to the susceptibility is found to be , _
corr^-1&=&-m_q p_f^3 .
the total susceptibility is given by ^nr&=&_p^-1 . 0.2 in in fig.([chi_comp ] ) we plot inverse spin susceptibility which is valid for all the kinematic regimes .
it shows @xmath2 changes its sign at the density @xmath50 without correlation correction and when we include the correlation effect its sign changes at @xmath51 .
this is equivalent to what happens to the ground state energy as a function of @xmath18 .
it is needless to mention that this change of sign correspond to the para - ferro phase transition in dense quark system .
the parameter set used here are same as those of @xcite .
once we have the expressions for the total energy density , the incompressibility @xmath52 and sound velocity @xmath53 can be determined .
the incompressibility @xmath4 is defined by the second derivative of the total energy density with respect to the number density @xmath54 , which is given by@xcite k&=&9n_q . since , there are two fermi surfaces corresponding to spin - up @xmath55 and spin - down @xmath56 states , such that @xmath57 .
we have @xcite [ dele_deln ] & = & + + & = & .
the single particle energy at the fermi surface or chemical potential of spin - up quark turns out to be [ up_chpot ] ^+,ur&=&_kin^++_ex^++_corr^+ + & = & p_f^++(p_f^++ ) + ( p_f^++ ) .
similarly , @xmath58 can be obtained by replacing @xmath59 with @xmath60 in eq.([up_chpot ] ) . in ref.@xcite
, chemical potential is determined within fermi liquid theory approach upto @xmath7 .
we , however , here calculate @xmath61 with different approach upto @xmath62 . using eq.([dele_deln ] ) and eq.([up_chpot ] ) , the incompressibility becomes [ incomp ] k^ur&=&p_f\{[(1+)^4/3+(1-)^4/3 ] + [ ( 1+)^4/3+(1-)^4/3 + 2(1-^2)^2/3 ] . + & & .
+ [ ( 1+)^4/3+(1-)^4/3 + 2(1-^2)^2/3]}. another interesting quantity would be to calculate the first sound velocity which is given by the first derivative of pressure with respect to energy density .
mathematically @xcite , c_1 ^ 2&=&. from eq.([up_chpot ] ) , we have & = & \{1 + .
+ & & . + } .
the second and last term in the curly braces corresponds to exchange and correlation contribution respectively .
similarly , @xmath63 can be obtained by replacing @xmath18 with @xmath64 .
using @xmath65 , @xmath61 and @xmath66 , we calculate the sound velocity in terms of @xmath18 .
numerically , for unpolarized matter @xmath67 , while for complete polarized matter @xmath68 , which is below the causal value @xmath69 at the high density limit . 0.2 in in fig([compres ] ) , we plot the density dependencies of the incompressibility with correlation correction .
this shows for higher value of the order parameter @xmath18 , the incompressibility becomes higher for the same value of density .
thus numerical values of incompressibility and sound velocity shows that equation of state for polarized quark matter is stiffer than the unpolarized one @xcite .
in this section we calculate the exchange energy density @xmath22 at low - temperature @xmath70 , for which we replace @xmath71 of eqs.([ex_nf]-[ex_f ] ) with proper fermi distribution function . in the ultra - relativistic limit
, the angular averaged interaction parameter is given by @xcite [ int_ur ] f_pp^ur&= & the spin non - flip contribution to the exchange energy density is [ exnf_ur ] e_ex^nf&=&_s= f_pp^nf n_p^s(t ) n_p^s(t ) + & & p_f^4+t^2p_f^2 .
here @xmath72 is the fermi distribution function .
similarly , @xmath73 can be evaluated .
the total @xmath74 at low temperature is found to be [ ex_urt ] e_ex^ur&&p_f^4 + & + & t^2p_f^2 .
the kinetic energy density can be written as [ kin_urt ] e_kin^ur & & + . from eq.([xi_expan ] ) each energy contribution to the susceptibility is _
kin^-1&=&(1- ) + _ ex^-1&=&-(1 + ) . [ exsus_t ]
it is to be noted that the @xmath75 independent terms of the above expressions are identical with those eqs.([ur_eng]-[ur_contri ] ) .
thus the susceptibility at non - zero temperature is given by [ sus_urt ] ^ur&=&_p^-1 .
in the non - relativistic limit the interaction parameter takes the following form @xcite f_pp^nr=- . for spin anti - parallel interaction @xmath46 , then @xmath76 .
thus the contribution due to the scattering of quarks with unlike spin states vanishes and the dominant contribution to energy density comes from the parallel spin states @xmath25 . performing the angular integration of eq.([ex_nf ] )
, the exchange energy density upto term @xmath77 becomes e_ex^nr&=&-_s = p dp n_p^s(t ) p dp n_p^s(t)|| + & & -p_f^4-t^2 m_qp_f .
the kinetic energy density is found to be e_kin^nr & & + .
separate contribution from kinetic and exchange energy to susceptibility becomes _ kin^-1&= & ( 1- ) + _ ex^-1&=&- ( 1- ) .
thus , at low temperature the susceptibility turns out to be ^nr&=&_p^-1 .
in this work we have derived the spin susceptibility for degenerate quark matter with corrections due to correlation contributions .
analytic expressions for susceptibility are also derived both in the ultra - relativistic and the non - relativistic limit . it is observed that at low density susceptibility changes sign and becomes negative suggesting the possibility of ferromagnetic phase transition .
in addition , we also derive single particle energy , sound velocity and incompressibility upto @xmath62 . as far as the equation of state is concerned , in the present model
, we find that equation of state for polarized matter is stiffer than that of unpolarized one .
we also determine the exchange energy and susceptibility at non - zero temperature of the spin polarized quark matter .
to calculate the correlation contribution to the spin susceptibility we have from eq.([cor_expan ] ) _ 1l & = & - + & & \ { } + & & \{64_f^5 ^ 2_e^-1(v_f_e ) ( p_f^2+m_q^2 ^ 2_e)^2 + & - & 2p_f_f^22_e^-1(v_f_e ) \{12m_q^6 + 51m_q^4p_f^2 + & + & m_q^4(4m_q^2 + 5p_f^2)4_e + 68m_q^2p_f^4 + 4m_q^2(4m_q^4 + 10m_q^2p_f^2 + 7p_f^4)2_e + 32p_f^6 } + & + & 4p_f^2_f^2_e[6m_q^6 + 29m_q^4p_f^2 + m_q^4(2m_q^2 + 3p_f^2)4_e+36m_q^2p_f^4 + & + & 4m_q^2(2m_q^4 + 4m_q^2p_f^2 + 3p_f^4)2_e+16p_f^6 ] } [ a_1l ] + b_1 t & = & + & & \ { } + & & \{-32_f^3 ^ -1(v_f_e ) ( m_q^2 ^ 2_e+p_f^2)^2 + & - & 8p_f^2_f[m_q^4+p_f^4+m_q^2p_f^2(1+^2_e ) ]
^2 2_e+2p_f^-1(v_f_e)2_e + & & [ 8m_q^6 + 31m_q^4p_f^2+m_q^4p_f^24_e+36m_q^2p_f^4 + 4m_q^2(2m_q^4 + 4m_q^2p_f^2 + 3p_f^4)2_e+16p_f^6]}. + [ b_1 t ] 0.2 in * acknowledgments * + the authors would like to thank s.mallik for the critical reading of the manuscript and s.biswas for his help with the numerical calculations .
the author k.pal would like to special thanks t.tatsumi for his valuable suggestions .
99 t.tatsumi , phys.lett.*b 489 * , 280 ( 2000 ) .
f.bloch , z.phys.*57 * , 545 ( 1929 ) .
a.niegawa , prog.theor.phys * 113 * , 581 ( 2005 ) . s.a.chin and a.k.kerman , phys.rev lett . * 43 * , 1292 ( 1979 ) .
e.nakano , t.maruyama and t.tatsumi , phys.rev.*d68 * , 105001 ( 2003 ) .
k.ohnishi , m.oka and s.yasui , phys.rev.*d76 * , 097501 ( 2007 ) .
m.inui , h.kohyama and a.niegawa , arxiv:0709.2204 . t.tatsumi and k.sato , phys.lett * b663 * , 322 ( 2008 ) .
t.tatsumi and k.sato , phys.lett * b672 * , 132 ( 2009 ) .
k.sato and t.tatsumi , nucl.phys .
* a826 * , 74 ( 2009 ) k.pal , s.biswas and a.k.dutt-mazumder , phys.rev.*c79 * , 015205 ( 2009 ) . b.a.freedman and l.d.mclerran ,
phys.rev.*d 16 * , 1130 ( 1977);*16 * , 1147 ( 1977);*16 * , 1169 ( 1977 ) .
s.a.chin , ann.phys . * 108 * , 301 ( 1977 ) .
j.i.kapusta , _ finite temperature field theory _ , ( cambridge university press ) , ( 1989 ) .
a.l.fetter and j.d.walecka , _ quantum theory of many - particle systems _ , ( dover publications , inc . new york . )
k.pal , s.biswas and a.k.dutt-mazumder , phys.rev.*c80 * , 024903 ( 2009 ) .
m.gell-mann and k.a.brueckner , phys.rev.*106 * , 364 ( 1957 ) .
k.sawada , k.a.brueckner and n.fukuda , phys.rev.*108 * , 507 ( 1957 ) .
p.nozires and d.pines , phys.rev.*111 * , 442 ( 1958 ) .
d.pines , _ the many - body problem _ , lecture note , ( new york , 1961 ) .
j.p.perdew and y.wang , phys.rev.*b45 * , 13244 ( 1992 ) .
k.a.brueckner and k.sawada , phys.rev.*112 * , 328 ( 1958 ) .
b.s.shastry , phys.rev.lett.*38 * , 449 ( 1977 ) .
b.s.shastry , phys.rev.*b17 * , 385 ( 1978 ) .
f.perez , arxiv : 0810.4872 .
c.herring , _ magnetism _ , edited by g.t.rado and h.shul ( academic press , new york , 1966 ) , vol . | the expression for the spin susceptibility @xmath0 of degenerate quark matter is derived with corrections upto @xmath1 .
it is shown that at low density , @xmath2 changes sign and turns negative indicating a ferromagnetic phase transition . to this order , we also calculate sound velocity @xmath3 and incompressibility @xmath4 with arbitrary spin polarization .
the estimated values of @xmath3 and @xmath4 show that the equation of state of the polarized matter is stiffer than the unpolarized one .
finally we determine the finite temperature corrections to the exchange energy and derive corresponding results for the spin susceptibility . |
dikpati et al ( 2006 ) first used a flux transport dynamo calibrated to the sun ( dikpati et al 2004 ) to simulate and predict solar cycle peaks from the record of past surface magnetic field patterns .
this was done mathematically by forcing the dynamo equations at the top boundary , with a forcing function derived from past surface magnetic fields .
flux transport dynamos , and indeed all dynamos , have their own unforced , usually complex frequencies of excitation that are commonly found by treating the dynamo equations as an eigenvalue problem .
many naturally occurring and man - made systems have such properties .
when a physical system that has natural freqencies is excited by external forcing whose own frequency is close to one of the natural ones , there can be resonance produced that is , the system will be excited strongly by the forcing compared to the case where the forcing frequency is not close to a natural one .
the purpose of this paper is to explore the possibility of resonance in flux - transport dynamos relevant to the solar cycle . in flux transport dynamos ,
there are several physical properties that help determine the unforced frequencies of the system .
these include differential rotation , meridional circulation , the so - called @xmath0-effect , or kinetic helicity , and turbulent magnetic diffusion .
it is now well established ( dikpati and charbonneau , 1999 ) that unless the magnetic diffusivity is very large , meridional flow at the bottom of the dynamo layer is primarily responsible for the real part of the natural frequency of the dynamo , which determines the speed with which induced toroidal and poloidal fields near the bottom migrate toward the equator . therefore the closeness of the frequency of forcing at the top to the speed of the flow at the bottom could help determine how much dynamo response there is . since the forcing at the top is created by emergence of concentrated magnetic flux from the bottom , in the form of active regions , and the rate of movement of the zone where active regions are found moves toward the equator ( not coincidentally ) at a rate close to the meridional flow speed near the bottom , we might expect the conditions for resonance to occur in the bottom layer to be favorable . on the other hand , we know from observations ( ulrich , 2010 and references therein ) that the meridional flow at the top of the convection zone is toward the poles , opposite to the propagation of the surface forcing as well as 5 - 10 times faster .
thus we should not expect resonance to occur near the surface .
it is also well known ( ulrich 2010 and references therein ) that the meridional circulation varies with time .
this time variation is now being incorporated into a flux - transport dynamo used for prediction by dikpati and colleagues . in the 2006 prediction ,
meridional circulation generally was kept fixed in time .
dikpati et al ( 2006 ) , dikpati and gilman ( 2006 ) recognized that such time variations could be important , but felt they lacked sufficient knowledge of its variations to include them .
they adjusted the time - independent meridional flow amplitude to give the average period of the past solar cycles , and stretched or compressed all the surface forcing data to the same period , to avoid any artificial or non - physical mismatches between the natural dynamo period and the period of the forcing .
but there can also in principle in the sun be real differences between the period of the top forcing that was created by the previous cycle , and the freqency of equatorward propagation associated with the meridional flow speed at the bottom . in dynamos forced at the top with a specified period
, the amplitude of the induced fields within the dynamo domain will be affected by this frequency difference .
the model we present here in effect studies how this amplitude is affected , by treating the meridional flow at the bottom as a free parameter while keeping the frequency of the top forcing fixed . in the real sun , the cycle period varies from cycle to cycle , as does the speed of the meridional flow and its profile with latitude .
ultimately it is highly desirable to include both such variations .
this can be done by use of data assimilation techniques applied to both the surface forcing and meridional flow variations . as we said above , dikpati and colleagues
are doing that now .
when that is accomplished , they may find that resonance plays some role .
in this paper , we anticipate that possibility and focus on possible resonances by using a much simpler dynamo model than used in dikpati and gilman ( 2006 ) , namely one that has no more than two three layers in the radial direction .
such an approach has the advantage of speed while retaining important physical processes .
but such a simple model would have little value as a tool for prediction , because it could not be calibrated well in detail to the sun , since it would have few degrees of freedom .
it also may overestimate the importance of resonance for the same reason .
the cautions expressed in roald ( 1998 ) about the limits of dynamo models with one or two layers are well taken .
nevertheless , since the forced dynamo problem has only begun to be studied , particularly in the solar case , using a really simple model initially may give useful guidance about what to look for with a more realistic version .
it is in this spirit that we report on these calculations here .
resonance has been studied in dynamos previously , but the literature is small .
general examples include strauss ( 1986 ) and reshetnyak ( 2010 ) .
resonance in the geodynamo has been studied by stefani and gerberth ( 2005 ) and fischer et al ( 2008 ) .
studies for disks and galaxies include chiba ( 1991 ) , schmitt and rdiger ( 1992 ) , kuzanyan and sokoloff ( 1993 ) , and moss ( 1996 ) .
we have not located any previous studies specific to the sun in which resonance has been explicitly identified and highlighted . however in all these areas there are almost certainly model studies in which some form of resonance is playing a role , but has not been brought out in the analysis of results .
any dynamo in which inputs such as flow fields or turbulent parameters such as the @xmath0-effect are allowed to vary with time , either imposed or by nonlinearities internal to the system , such as quenching , could display behavior related to resonance .
we start from the standard flux - transport dynamo equations in vector form that include differential rotation , meridional circulation , turbulent magnetic diffusivity , and allow for an inhomogeneous top boundary condition . in vector - invariant form , this equation is given by @xmath1 in this equation , @xmath2 is the vector magnetic field , @xmath3 is the vector velocity , @xmath0 is the well known alpha - effect , and @xmath4 is the turbulent magnetic diffusivity . in addition , @xmath2 must satisfy the divergence - free condition @xmath5 .
this can be a problem numerically if we are solving for the three vector components of the magnetic field directly .
but satisfaction of this condition is guaranteed if instead we define the magnetic field in terms of vector potential functions
. there are at least two possible ways to do this .
the standard one from electromagnetic theory is to let @xmath6 , factor out a curl operator from the whole equation .
the resulting equation is given by @xmath7 equations ( 1 ) and ( 2)are quite general , including parameters and variables that can vary with all three space dimensions , in any coordinate system .
here we simplify the problem to an infinite plane layer and use cartesian geometry .
we restrict the system further by assuming all quantities are independent of one coordinate in the plane , which we take to be the @xmath8 coordinate .
we identify this coordinate with longitude on the sun , so that in the cartesian system the @xmath9 plane corresponds to the meridional plane on the sun .
the coordinate @xmath10 then corresponds to colatitude and @xmath11 to radius .
then in this frame , we take the velocities to be @xmath12 in the @xmath13 directions respectively . to describe the magnetic field in this system
requires only two variables , namely the @xmath8 components of the toroidal field and the poloidal potential .
we denote these scalar quantities by @xmath14 and @xmath15 respectively .
the magnetic diffusivity is denoted by @xmath4 .
then the vector system in equations ( 1 ) and ( 2 ) reduces to a pair of scalar equations for @xmath15 and @xmath14 as follows @xmath16 @xmath17 @xmath18 we simplify the problem further by restricting all coefficients in equations ( 3 ) and ( 4 ) to be independent of @xmath10 .
this allows for separation of variables . to achieve this
we must require @xmath19 so that @xmath20 can be independent of @xmath10 , and we must allow @xmath21 to be either a function of @xmath10 ( latitude ) only , or a function of @xmath11 ( radius ) only . there is evidence ( dikpati et al 2005 ) that the latitude gradient of rotation is more important than the radial gradient in the flux transport dynamos that best simulate solar cycles , so in this study we restrict ourselves to consideration of the latitude gradient .
the diffusivity @xmath4 and the @xmath0-effect are also taken to be independent of @xmath10 .
equations ( 3 ) and ( 4 ) then reduce to @xmath22
@xmath23 @xmath24 there are a variety of boundary conditions that could be chosen for the top and bottom of the infinite plane layer .
consistent with solar conditions , we take the bottom to be a perfect conductor , and therefore require @xmath25 there .
@xmath14 in that case is determined internally . for the top
there are four plausible alternatives : perfect conductor ( @xmath25 again);insulator with no forcing ( @xmath26 and @xmath15 matched to potential field above ) ; forcing in potential at top ( @xmath27 ) and @xmath26 or determined internally .
we make choices among these boundary conditions when we derive the 1- and 2-layer equations in the next section . in preparation for these derivations , we simplify the dynamo equations further by taking @xmath28 all independent of x , and assuming all variables have solutions of the form @xmath29 equations ( 5 ) and ( 6 ) reduce to @xmath30 @xmath31
before developing 1 , 2 and 3-layer dynamo equations in detail , we first describe schematically what variables and parameters are retained in these cases .
these are summarized graphically in figure 1 .
for all three cases , it is possible to specify boundary values of the variables in addition to their values within each layer .
we denote values at the top by a subscript @xmath32 and at the bottom by a subscript @xmath14 .
the boundary conditions shown correspond to a perfectly conducting bottom and an insulating or vacuum top with poloidal forcing @xmath33 . in the 2 and 3-layer cases
, we can also specify values of the variables at the interface between the upper and lower layers , namely the average of the values above and below . in some studies
, the variables are allowed to vary with the vertical coordinate within the layer . here
they vary in the vertical only from layer to layer .
some readers may prefer the term level to that of layer to describe our model . on the left of figure 1
are shown the variables @xmath34 and the parameters @xmath35 for each layer in the model , denoted by the subscripts @xmath36 for three layers , @xmath37 for two layers , and unsubscripted for the 1-layer version .
the whole depth of the dynamo domain is taken to have a thickness h , no matter how many layers it is subdivided into .
we now use the boundary conditions defined in figure 1 to evaluate the vertical derivatives in equations ( 7 ) and ( 8) . with rearrangements to put only forcing terms on the right hand sides , these equations become @xmath38a-{\alpha}b={4{\eta } \over h^2}a_f\quad\eqno(9)\ ] ] @xmath39b= -({4{\alpha}\over h^2}+{s\over h})a_f\quad\eqno(10)\ ] ] we can reduce the number of parameters to take account of by making equations ( 9 ) and ( 10 ) dimensionless , by using @xmath40 as the length scale , @xmath41 as the time scale and recognize that @xmath42 scales as @xmath40 . then velocities and @xmath0 scale as @xmath43 and equations ( 9 ) and ( 10 ) become @xmath44a-{\alpha}b=4a_f\quad\eqno(11)\ ] ] @xmath45b=-(4{\alpha}+s)a_f\quad\eqno(12)\ ] ] we find solutions to equations ( 11 ) and ( 12 ) from standard linear equation theory . in the case with no forcing , there are solutions only if the @xmath46 determinant @xmath47 of the coefficients vanishes .
this requires that both the real and imaginary parts of @xmath47 be zero .
dt is given by @xmath48[-i(\omega - kv)+k^2 + 4]-{\alpha}^2(k^2 + 8)=0,\quad\eqno(13)\ ] ] this yields a solution for @xmath49 in terms of the mode wavenumber @xmath50 and the meridional flow @xmath20 and @xmath0-effect , given by @xmath51^{1/2})\quad\eqno(14)\ ] ] when the factor multiplying @xmath52 in equation ( 14 ) is real , then ( 14 ) yields modes that propagate at the speed of the meridional flow , and either grow or decay with time .
one growing mode is assured if @xmath53 , so if the latitudinal scale of the mode is sufficiently large , a growing dynamo mode results . this is a form of flux - transport dynamo , but one in which differential rotation plays no role , even though it is present .
this is true only of the one - layer model ; with two or more layers , differential rotation does play a role in the unforced dynamos .
when @xmath54 then there are either all decaying modes , or oscillatory modes that propagate at speeds different from the meridional flow speed .
we can think of the condition @xmath55 as the threshhold for dynamos to occur . in the case with forcing @xmath33 at ( real )
frequency @xmath56 , bounded solutions to equations ( 11 ) and ( 12 ) exist only if the determinant @xmath47 does not vanish . in this case , the solutions for @xmath14 and @xmath15 in terms of @xmath33 are given by @xmath57 @xmath58\quad\eqno(16)\ ] ] > from equation ( 15 ) for the toroidal field @xmath14 we can see immediately that the toroidal field induced in response to the poloidal forcing @xmath33 is largest when @xmath47 is smallest , and is unbounded when @xmath59 .
therefore the system experiences resonance , when the frequency of the forcing equals the frequency @xmath60 associated with the meridional flow and the wavenumber of the forcing equals@xmath61 .
note that this is the same point in parameter space where there are neutral dynamo waves as found from equation ( 13 ) .
just as in resonant systems generally , here the natural frequency of the system and the frequency of the forcing are the same .
so the presence of resonance requires an @xmath0-effect , but from equation ( 15 ) it does not require a differential rotation .
the resonance in this 1-layer model is independent of the sign of the @xmath0-effect .
it is well known that @xmath62 dynamos can have propagating unstable dynamo modes , so resonance is possible in such systems without the effect of differential rotation . in models with two or more layers , resonance does involve differential rotation . what is the relevance of this resonance phenomenon to the solar cycle ?
we know that at the top of the convection zone , the meridional flow and surface poloidal forcing are propagating in opposite directions , with the poleward meridional flow being an order of magnitude greater .
therefore in this domain the dynamo should be far from resonance .
but at the bottom , the meridional flow is toward the equator , and has a speed similar to the propagation speed of the surface poloidal source . therefore if the surface poloidal source can be felt near the bottom of the convection zone
, resonance might occur .
a 1-layer dynamo model is inadequate to test for this possibility , so we must use a model with at least two layers .
for the 2-layer model equations , we proceed in the same way as with the 1-layer equations . but here , as listed in figure 1 , most parameters have different values for the upper and lower layers .
but for simplicity and to be able to achieve separation of variables , we keep the @xmath8 ( latitudinal ) gradient @xmath63 of the east - west ( rotational ) flow the same in both layers . then the dimensional 2-layer equations become @xmath64a_u-{\alpha}_ub_u-2{\eta}_ua_l/{h_u}^2=4{\eta}_ua_f/{h_u}^2\quad\eqno(17)\ ] ] @xmath65b_u+(2{\alpha}_u/{h_u}^2-s/2h_u)a_l\ ] ] @xmath66 @xmath67a_l-{\alpha}_lb_l=0\quad\eqno(19)\ ] ] @xmath68 @xmath69a_l\ ] ] @xmath70b_l=0\quad\eqno(20)\ ] ] by inspection , we can see that the layers are coupled by processes involving diffusion , differential rotation and the @xmath0-effect . to render the system dimensionless in this case
, we use @xmath40 for the length scale and @xmath71 for the time scale . we also introduce the parameters @xmath72 respectively for the ratios @xmath73 .
we also define @xmath74 and @xmath75
. then @xmath76 . with these definitions and
evaluating the @xmath11 derivatives in equations ( 7 ) and ( 8) , we get four equations for the toroidal field and poloidal potential of the two layers . in dimensionless form
these equations are @xmath77a_u-{\alpha}_ub_u-2a_l/{d_u}^2=4a_f/{d_u}^2\quad\eqno(21)\ ] ] @xmath78b_u+(2{\alpha}_u/{d_u}^2-s/2d_u)a_l\ ] ] @xmath79 @xmath80a_l - p_{\alpha l}{\alpha}_ub_l=0\quad\eqno(23)\ ] ] @xmath81 @xmath82a_l\ ] ] @xmath83b_l=0\quad\eqno(24)\ ] ] by analogy with the 1-layer system , in the 2-layer case we should expect resonance to be found near where the ( 4x4 ) determinant of the coefficients in equations ( 17)-(20 ) vanishes . in the homogeneous case ( @xmath84 ) this determinant is a quartic equation for the complex eigenfrequency @xmath49 of the dynamos of the 2-layer system .
being a 4th order system , we can not in general find closed or simple algebraic forms for either the amplitudes of the response to the top forcing , or the phase speed and growth rate of the unforced dynamo .
but only small programs are necessary to get results . for the case with forcing , we find the amplitudes and phases of @xmath85 in terms of the amplitude and phase of @xmath33 by application of cramers rule ( refs ) .
we apply cramers rule to equations ( 21)-(24 ) defined in symbolic form as @xmath86 @xmath87 @xmath88 @xmath89 in which all of the @xmath90 coefficients and @xmath91 are defined by matching terms in equations ( 25)-(28 ) with their counterparts respectively in equations ( 21)-(24 ) , so that @xmath92 and @xmath93 . then to make cramers rule work , we must have the determinant of the coefficients in equations ( 25)-(28 ) not vanish .
but where it approaches zero is where in the parameter space we should expect resonance to occur , since this determinant is the denominator for the solutions of equations ( 25)-(28 ) found by cramers rule .
how would the results we have obtained change if we added another layer to the system ? if the results concerning resonance are similar , it gives us more confidence that the same phenomenom may occur in much more realistic systems with many layers in the vertical . if not , then the results obtained above have much more limited significance for the general case . with three layers
, the number of equations to be solved expands to six .
these are given by @xmath94a_u-\alpha_ub_u-{2 \over d_u^2}a_m={4 \over d_u^2}a_f\quad\eqno(29)\ ] ] @xmath95a_u+[-i(\omega - kv_u)+k^2+{6 \over d_u^2}]b_u+({2\alpha_u \over d_u^2}-{s \over 2d_u})a_m\ ] ] @xmath96 @xmath97a_m-\alpha_up_{\alpha m}b_m-{2p_{\eta m } \over d_m^2}a_l=0\quad\eqno(31)\ ] ] @xmath98a_u-{2p_{\eta m } \over d_m^2}b_u-\alpha_up_{\alpha m}(k^2+{4 \over d_m^2})a_m\ ] ] @xmath99b_m + [ { 2\alpha_up_{\alpha m}\over d_m^2}-{s \over 2d_m}]a_l-{2p_{\eta m } \over d_m^2}b_l=0\quad\eqno(32)\ ] ] @xmath100a_l-\alpha_up_{\alpha l}b_l=0\quad\eqno(33)\ ] ] @xmath101a_l\ ] ] @xmath102b_l=0\quad\eqno(34)\ ] ] similar to equations ( 25)-(28 ) above , equations ( 29)-(34 ) can be written in symbolic form as @xmath103 @xmath104 @xmath105 @xmath106 @xmath107 @xmath108 in which , as before , the @xmath109 coefficients are defined by matching terms from equations ( 35)-(40 ) respectively with those of ( 29)-(34 ) . but here there are many more coefficients that are zero , namely @xmath110 .
this greatly reduces the number of symbolic multiplys needed to apply cramers rule , rendering it practial to work out all the algebra for this 6x6 system .
equations ( 21)-(24 ) for the 2-layer model contain nine free parameters in addition to the poloidal forcing @xmath33 ; equations ( 29)-(34 ) for the 3-layer model contain 16 parameters .
therefore we must make judicious choices of parameter values . in making these choices
we will be guided by solar conditions as well as the uncertainties in the solar properties that define these parameters .
for example , the dimensionless frequency @xmath49 of the top forcing should be approximately the frequency of the solar cycle , corresponding to a period of 22 years .
with @xmath111 our frequency scale , for @xmath112 and @xmath113 this frequency is @xmath114 .
the solar cycle frequency is @xmath115 , so the dimensionless forcing frequency should be about 1.8 units .
therefore a frequency range of 1.5 to 2 . would cover most variability in solar cycles . in all calculations displayed below ,
we have chosen a dimensionless frequency @xmath49= 1.8 .
specifying the latitudinal wavemumber of the forcing is more uncertain .
the width of the sunspot zone in one hemisphere is about 30 degrees latitude , or @xmath116 .
this would be the minimum half wavelength of the forcing , but that forcing is seen to be broader in latitude scale than that , due to the dispersal and decay of active regions .
also , we never see surface fields from more than 2 sunspot cycles at the same time , so a more reasonable wavelength might be @xmath117 , the distance between equator and pole . then this wavelength would correspond to a dimensionless wavenumber @xmath118 units at the surface and @xmath119 units at the depth of the tachocline . an average value would be about 1.4 units , which is what we use for all calculations .
as for velocities , the velocity scale @xmath120 is about 1m / sec , so a typical solar meridional flow near the top would be 15 units , and the latitudinal differential rotation linear velocity relative to the rotating frame of about @xmath121 units .
we will use these dimensionless values to guide our choices of parameter ranges to survey .
for some purposes the choice of @xmath122 may be too high . reducing it by a factor of ten means that all dimensionless solar frequencies and velocities are increased by a factor of ten , but dimensionless wavenumbers remain the same .
the 1-layer results given above could give us guidance about where to look in parameter space for resonance when there are more layers than one .
as in that case , we might expect that for resonance to occur in a layer , we must have the phase speed of the forcing at the top be equal to or vary close to the meridional flow speed in that layer .
we have established both algebraically and numerically that resonance does happen in the bottom most layer of the system when the meridional flow satisfies that condition and the cross product of the coefficients of @xmath123 and @xmath124 approaches zero . in terms of formulas , this resonance occurs in the 2-layer model in the neighborhood of @xmath125 and in the 3-layer model near where @xmath126 in both 2 and 3-layer models this implies @xmath127 and @xmath128 the fact that the conditions for resonance are identical in the 2 and 3-layer cases suggests that by induction that this will remain true no matter how many layers the model contains .
therefore it is likely to be a robust general property of this flux - transport dynamo , but we have not attempted to prove this mathematically .
it is evident from equation ( 44 ) that the @xmath0-effect in the bottom layer plays an important role in creating resonance there .
some flux - transport models applied to the sun contain no @xmath0-effect there , but dikpati and gilman ( 2001 ) showed that its presence could be responsible for choosing the correct symmetry for the sun s toroidal and poloidal fields ( see also bonanno et al 2002 , hotta & yokoyama 2010 ) .
the resonance we demonstrate in this work gives further importance to knowing what the @xmath0-effect is at the base of the convection zone .
the conditions ( 43 ) and ( 44 ) guarantee an essentially infinitely large resonance ( interestingly even though there is diffusion in the problem , which usually bounds the resonance to a finite value ) , but to be realized requires a precise combination of values of several parameters of the problem , very unlikely to be realized . but just being close to resonance is enough to increase the response of the system to the same forcing at the top by a factor of 10 - 100 , beyond the range of variation in solar cycle peaks .
so it is worth mapping out the response over a wide range of parameter values that are plausible for the sun .
but at the same time we must recognize that this 3-layer model is much simpler than the real sun , and much simpler than 2d flux transport dynamos in spherical shells .
solutions of equation ( 44 ) in the limit of small @xmath129 are of particular interest , since we expect the bottom layer to have the lowest magnetic diffusivity . in that limit
the two roots are @xmath130 . in that case
, there is resonance even if the @xmath0-effect is zero in the bottom layer .
here we answer the question of where in the parameter space we will find resonance occurring . in organizing the results it is helpful to differentiate roles played by the various parameters in the real sun , including how they might vary with either time or space .
for example , while we do not know with precision the thickness of the bottom layer of the convection zone where we expect the magnetic diffusivity to be small , this thickness should not vary much with time , or probably with latitude .
therefore we can think of the thickness as an externally specified parameter .
so we would like to know where in the range of other parameters resonance should occur , for a selection of assumed bottom layer thicknesses .
similarly , the solar differential rotation does not appear to vary much with time in the sun ( the well - known torsional oscillations in rotation are about one - half of one percent of the equatorial rotation ) , so we can treat it in the same way .
the meridional flow varies more with time , but because of the density increase with depth , we expect the flow in the lower layer always to be much less than that of the upper layer and in the opposite direction .
somewhat arbitrarily we take the ratio of the two , @xmath131 .
we also take the meridional flow of the middle layer to equal that of the upper layer , for simplicity and with some guidance from observations , e.g. gizon and rempel ( 2008 ) , that do not show a reversal in meridional flow at mid - depth in the solar convection zone .
> from theoretical considerations , we also do not expect the magnetic diffusivity in the bottom layer to vary with time , since it represents an average over many turbulent fluctuations ; we do expect it to vary with depth , and this we have taken into account within the limitations of a 2 or 3 layer model .
by contrast , the @xmath0-effect in the bottom layer may come from the action of a relatively small number of global events , such as global mhd instability , so it might fluctuate greatly with time , perhaps even changing sign .
therefore we choose to display our results as plots of the @xmath0-effect needed for resonance as a function of the other parameters .
figure 2 displays @xmath132 needed for resonance as a function of the magnetic diffusivity of the lower layer , for a solar differential rotation ( @xmath121 ) and a selection of lower layer thicknesses ( the upper layer thickness is held fixed at @xmath133 of the total thickness ) . since the y - axis is logarithmic , we have reversed the sign of all the negative @xmath132 values . viewed from the left hand @xmath8-axis , the positive @xmath132 values are the upper family of curves , while the negative @xmath132 values are the lower family .
the negative values are all asymptotic to zero with decreasing @xmath129 , while the positive values are all asymptotic to constant nonzero values , consistent with the analytical limits shown in section 6.1 . for increasing values of @xmath129 the positive and negative @xmath132 values for the same lower layer thickness are asymptotic to the same amplitudes , because in that limit the effect of differential rotation becomes insignificant .
the effect of differences for different lower layer thicknesses also become less , because as the thickness increases , vertical diffusion shrinks to an amplitude closer to the latitudinal diffusion , which is proportional to @xmath134 , the square of the latitudinal wavenumber of the externally imposed forcing .
figure 2 shows that for all lower layer magnetic diffusivities , there is always at least one @xmath135 available to create resonance , and two available if the lower layer diffusivity is high enough .
this is true for all lower layer thicknesses shown , which cover the range of reasonable values for the thickness of the layer with the lowest magnetic diffusivity in the dynamo domain @xmath136 .
a reasonable assumption for the sun is that the turbulent magnetic diffusivity at the bottom is a factor @xmath137 smaller than at the top . with our scaling and the curves shown in figure 2 ,
this means resonance should occur for dimensional @xmath132 values in the range @xmath138 , which are very typical values used in flux - transport dynamos applied to the sun .
resonance should also occur for much smaller values of @xmath132 of the opposite sign .
we can give some physical interpretation of the results shown in figure 2 and contained in the formulas used to generate it .
there is a competition among the physical processes associated with the @xmath0-effect , magnetic diffusion , differential rotation and meridional circulation to determine where in parameter space resonance will occur .
in general , diffusion works against resonance , while the @xmath0 effect and differential rotation work to produce resonance by inductive processes .
but , depending on their signs , the @xmath0-effect and differential rotation work with or against each other . from equation ( 44 ) , if the product @xmath139 is negative they work in concert , and if it is positive , they are in opposition .
we have taken @xmath63 positive for the differential rotation , so when @xmath140 they reinforce , and when @xmath141 they oppose each other .
what happens is that oppositely signed @xmath132 leads to oppositely directed poloidal potential and therefore oppositely directed toroidal fields induced by the same differential rotation .
all patterns are swept toward the equator by the meridional flow . in one case
the induced toroidal field on the leading edge of the pattern reinforces what is already present , and in the other case it tends to cancel it out .
clearly the former case leads to a stronger response , or approach to resonance .
this approach to resonance is optimized by choosing the meridional flow speed to match the forcing speed , effectively freezing the phase of the forcing relative to the phase of the induced fields , allowing for maximum amplification .
when @xmath132 and @xmath63 are working together , for a given @xmath63 , a smaller @xmath132 is needed to achieve resonance .
hence for all diffusivities , the negative @xmath132 values found for resonance are smaller in amplitude than their positive counterparts for the same values of other parameters , as seen in figure 2 .
the differential rotation value @xmath121 in our scaling is equivalent to the whole differential rotation of the sun with latitude at the photosphere .
but the lower layer of the model applies to the bottom of the convection zone and the tachocline , where the latitudinal differential rotation would be smaller by an amount that is a fairly strong function of depth .
how much difference in the values of @xmath132 that are needed to produce resonance occurs for lower differential rotation ?
figures 3 and 4 give the answer . in figures 3 and 4 we display @xmath132 needed for resonance as a function of the differential rotation parameter @xmath63 , for the same lower layer thicknesses as shown in figure 2 , for two selected lower layer diffusivities @xmath142 ( figures 3 , 4 respectively ) , corresponding to @xmath143 , that cover the range of plausible values for the bottom of the solar convection zone .
here we use a linear scale for @xmath132 so that we can retain its sign .
the primary message from figures 3 and 4 is that , for lower layer thicknesses and diffusivities plausible for the sun , resonance occurs for all differential rotations possible in the lower layer , for values of @xmath135 .
the positive @xmath132 needed for resonance increases with differential rotation @xmath63 , since in this case the @xmath0-effect and effect of differential rotation oppose each other , while the negative @xmath132 needed declines in amplitude with increase in @xmath63 , because in this case the two effects reinforce each other .
the results shown in figures 2 - 4 apply to both two- and three - layer models , since equations ( 43 ) and ( 44 ) contain quantities only from the lower layer .
but clearly the three - layer system captures more physics , so in the next section we focus on amplitude results for three layers .
the corresponding results for two layers are qualitiatively similar .
the previous subsection presented guidance for where in our parameter space to find resonance .
here we present what that resonance looks like as functions of the various parameters of the problem .
figures 5 and 6 display the amplitudes of toroidal field and poloidal potential for all three layers of the model for poloidal forcing @xmath144 for two selected lower layer magnetic diffusivities , the same as for figures 3 and 4 respectively .
these results are for full solar differential rotation ( @xmath121 ) , a lower layer thickness of @xmath145 of the total thickness , and a meridional flow speed at the top of @xmath146 , which is predicted from equation ( 43 ) for resonance in the lower layer for @xmath147 and @xmath148 when @xmath131 .
it is important to realize that the amplitudes shown in figures 5 - 7 are not exponentially growing as in the usual unforced dynamo solutions , but instead represent amplitudes of forced oscillatory solutions . strictly speaking , these are not self excited dynamos , because of the top boundary forcing , but they are dynamos nontheless . for these parameter values , equations ( 43 ) and
( 44 ) predict resonance will occur near @xmath149 for @xmath150 and @xmath151 for @xmath152 .
we see strong upward spikes in @xmath153 , the amplitude ( absolute value ) of the toroidal field in the lower layer ( gold curve ) at about these values of @xmath132 .
the lower layer poloidal potential @xmath154 ( red curve ) also peaks there .
smaller peaks at the same @xmath132 are present in the toroidal fields in the middle and upper layer ( @xmath155 , dark blue curve;@xmath156 , dark green curve ) while the poloidal potentials ( @xmath157 , light blue curve ; @xmath158 , light green curve ) of these layers respond to the resonance hardly at all .
finally , there is a downward spike in the poloidal potential of the lower layer at @xmath159 .
the poloidal potential of the upper layer is much larger than that of the middle layer because the former is determined directly by the forcing at the top , while in the middle layer only a small @xmath160 is present .
several features of figures 5 and 6 are notable .
we see in figure 5 that for all values of @xmath132 there is a large response in the lower layer , compared to the other layers , to the forcing applied at the top , even though the magnetic diffusivity in the lower layer is the same as that of the middle layer .
this is because the meridional flow and the @xmath0 of the middle layer are not close to the values needed for resonance in that layer .
but at the same time , the middle and upper layers do show successively lower but still significant peaks in toroidal fields near the same values of @xmath132 .
this is caused by magnetic diffusion upward across the interfaces between layers .
the main changes seen in figure 6 compared to figure 5 are that with diffusivity of the lower layer reduced by a factor of ten , the resonance becomes much narrower and sharper .
in other words , with lower diffusivity the range of @xmath132 over which there is substantial amplification of the effect of the forcing at the top is narrower .
but where resonance does occur , the middle and upper layers respond more strongly to the resonance for non - zero @xmath132 .
this is not true for the resonance near @xmath159 , because such a low value leads to less production of poloidal field in the lower layer ( compare the red curves in figures 5 and 6 in the neighborhood of the resonance for negative @xmath132 ) , from which the lower layer toroidal field must be produced by the differential rotation there .
figures 5 and 6 are for a rather precisely chosen lower layer meridional flow speed .
what happens to the resonance if we move away from that speed ?
we have examined this question by computing amplitudes of toroidal and poloidal fields as functions of @xmath132 for other speeds , namely @xmath161 .
we find the same resonances as seen in figures 5 and 6 , but with somewhat different amplification factors .
thus the presence of resonance of some significant amplitude is not strongly dependent on the precise value of the meridional flow .
this means in effect that equation ( 44 ) more closely determines the resonance than does equation ( 43 ) . in keeping with this inference ,
if we choose a value of @xmath132 only a few percent away from that predicted to give resonance , the resonance practically disappears no matter what speed of meridional flow is taken .
figure 7 displays the amplitude of the induced toroidal field in the lower layer as a function of meridional flow speed @xmath162 in the upper layer , for the @xmath132 values predicted for resonance for the same parameter choices as in figures 5 and 6 .
we see that for @xmath163 , the peak field does not occur at @xmath146 , at which equation ( 43 ) is satisfied , but rather at values above and below that ( blue and green curves , respectively ) , depending on the @xmath132 chosen .
while resonance still occurs for the predicted @xmath132 values , it is a factor of five to ten smaller than at the peaks shown .
this is again evidence that the closeness of equation ( 44 ) to being satisfied is the determining factor in closeness to resonance .
this is what makes the denominator of the algebraic expressions for @xmath153 and @xmath154 smallest .
this departure from the predicted points of resonance occurs because of the amplitude of the diffusivity assumed for the lower layer .
this is evident from figure 7 because the peak toroidal fields produced for @xmath152 ( red and yellow curves ) occur very close to the @xmath132 values predicted for resonance .
this also means that the resonant peaks shown in figure 5 are not the highest that are possible for @xmath164 , but the resonance is still quite pronounced for the parameter values chosen for figure 5 .
our principal conclusion is that , at least in this simple dynamo model , resonance will always be found , provided the right parameter values are chosen .
furthermore , it occurs for parameter choices that are plausible for the sun .
it occurs in the lowest layer of the model , where the meridional flow toward the equator is most likely to match , or nearly match , the propagation speed of the forcing imposed at the top , corresponding to the photosphere on the sun . and
the effect is large a factor of 10 - 100 amplification of toroidal field compared to the case where the parameter values are far from the ones for which resonance is predicted .
we acknowledge that this large a difference in amplitude for different parameter values is partly due to the model being kinematic .
if @xmath165 forces were included , the peaks would almost surely be smaller .
the solar convection zone is a spherical shell , not an infinite layer , so how well should we expect our results to apply to the sun ? in our model , the differential rotation @xmath63 is a constant , so the rotation is a linear function of latitude only , and independent of depth . in the sun , differential rotation varies with both latitude and depth . in our model ,
meridional flow is independent of latitude , and there is no vertical motion , while in the sun , meridional flow is a closed circulation confined by the boundaries of the convection zone as well as the poles .
meridional circulation is also known to vary with time on a variety of timescales .
the @xmath0-effect will also be a function of latitude as well as time .
these differences between our model and the sun imply that resonance , if it occurs , would likely be found in more localized regions of the dynamo domain .
but in principle it could still occur .
if a particular location experiences resonance or near resonance for some period of time , then the toroidal field might amplify quickly there , leading to locally anomalously large field .
how quickly presumably depends on the nearness to conditions for resonance , and how long these conditions last .
could such a sudden amplification be a precursor for the formation of a rising flux tube that creates a new active region ?
this seems like a hypothesis worth testing in the future .
it is also possible that in the convection zone conditions for resonance could persist for large fractions of a solar cycle and perhaps even from one cycle to the next .
if this happens , it could contribute to the evolution of the envelope of cycle amplitudes .
on extremely long time scales could persistent conditions favoring resonance be responsible for grand maxima , and conditions unfavorable for resonance lead to grand minima , such as the maunder minimum ?
clearly these are very speculative questions , requiring much more sophisticated solar dynamos than we have used here , to answer . in flux - transport dynamos in spherical shells that are forced from the top ,
it is the radial flow of the closed meridional circulation that transports the forcing signal including its frequency to the bottom . in our infinite plane cartesian model
, there is no radial flow , so the signal should be thought of as getting to the bottom locally via vertical diffusion and induction .
we can think of the meridional circulation in the upper layer closing at infinity and returning in the lower layer .
since we know that not all sunspot cycles have the same duration , the meridional flow will be bringing a frequency to the bottom that could be different than that implied by the meridional circulation there at the time the active regions were formed , that gave rise to the forcing frequency .
also , a change in the speed of the flow carrying the forcing signal will change the frequency of that signal by doppler - shifting it .
but in percentage terms the differences will not be large , so there could be continually evolving nearness to conditions needed for resonance , creating continuous changes in cycle amplitude , in addition to the usual evolution of a sunspot cyle through ascending , peak , declining and minimum phases . again , investigating such possibilities requires much more realistic dynamo models than we have used here . | we show that simple 2 and 3-layer flux - transport dynamos , when forced at the top by a poloidal source term , can produce a widely varying amplitude of toroidal field at the bottom , depending on how close the meridional flow speed of the bottom layer is to the propagation speed of the forcing applied above the top layer , and how close the amplitude of the @xmath0-effect is to two values that give rise to a resonant response .
this effect should be present in this class of dynamo model no matter how many layers are included .
this result could have implications for the prediction of future solar cycles from the surface magnetic fields of prior cycles .
it could be looked for in flux - transport dynamos that are more realistic for the sun , done in spherical geometry with differential rotation , meridional flow and @xmath0-effect that vary with latitude and time as well as radius .
because of these variations , if resonance occurs , it should be more localized in time , latitude and radius . |
weak value amplification ( wva ) @xcite is a concept that has been used under a great variety of experimental conditions @xcite to reveal tiny changes of a variable of interest . in all those cases ,
a priori sensitivity limits were not due to the quantum nature of the light used ( _ photon statistics _ ) , but instead to the insufficient resolution of the detection system , what might be termed generally as _
technical noise_. wva was a feasible choice to go beyond this limitation . in spite of this extensive evidence ,
its interpretation has historically been a subject of confusion " @xcite .
for instance , while some authors @xcite show that weak - value - amplification techniques ( which only use a small fraction of the photons ) compare favorably with standard techniques ( which use all of them ) " , others @xcite claim that wva does not offer any fundamental metrological advantage " , or that wva @xcite `` does not perform better than standard statistical techniques for the tasks of single parameter estimation and signal detection '' . however , these conclusions are criticized by others based on the idea that `` the assumptions in their statistical analysis are irrelevant for realistic experimental situations '' @xcite .
the problem might reside in here we make use of some simple , but fundamental , results from quantum estimation theory @xcite to show that there are two sides to consider when analyzing in which sense wva can be useful . on the one hand
, the technique generally makes use of linear - optics unitary operations .
therefore , it can not modify the statistics of photons involved .
basic quantum estimation theory states that the post - selection of an appropriate output state , the basic element in wva , can not be better than the use of the input state @xcite
. moreover , wva uses some selected , appropriate but partial , information about the quantum state that can not be better that considering the full state . indeed , due to the unitarian nature of the operations involved , it should be equally good any transformation of the input state than performing no transformation at all .
in other words , when considering only the quantum nature of the light used , wva can not enhance the precision of measurements @xcite .
on the other hand , a more general analysis that goes beyond only considering the quantum nature of the light , shows that wva can be useful when certain technical limitations are considered . in this sense , it might increase the ultimate resolution of the detection system by effectively lowering the value of the smallest quantity that can detected . in most scenarios ,
although not always @xcite , the signal detected is severely depleted , due to the quasi - orthogonality of the input and output states selected . however , in many applications , limitations are not related to the low intensity of the signal @xcite , but to the smallest change that the detector can measure irrespectively of the intensity level of the signal .
a potential advantage of our approach is that we make use of the concept of trace distance , a clear and direct measure of the degree of distinguishability of two quantum states .
indeed , the trace distance gives us the minimum probability of error of distinguishing two quantum states that can be achieved under the best detection system one can imagine @xcite .
measuring tiny quantities is essentially equivalent to distinguishing between nearly parallel quantum states .
therefore we offer a very basic and physical understanding of how wva works , based on the idea of how wva transforms very close quantum states , which can be useful to the general physics reader . here
were we use an approach slightly different from what other analysis of wva do , where most of the times the tool used to estimate its usefulness is the fisher information .
contrary to how we use the trace distance here , to set a sensitivity bound only considering how the quantum state changes for different values of the variable of interest , the fisher information requires to know the probability distribution of possible experimental outcomes for a given value of the variable of interest .
therefore , it can look for sensitivity bounds for measurements by including _ technical characteristics _ of specific detection schemes @xcite .
a brief comparison between both approaches will be done towards the end of this paper .
one word of caution will be useful here .
the concept of weak value amplification is presented for the most part in the framework of quantum mechanics theory , where it was born .
it can be readily understood in terms of constructive and destructive interference between probability amplitudes @xcite .
interference is a fundamental concept in any theory based on waves , such as classical electromagnetism .
therefore , the concept of weak value amplification can also be described in many scenarios in terms of interference of classical waves @xcite .
indeed , most of the experimental implementations of the concept , since its first demonstration in 1991 @xcite , belong to this type and can be understood without resorting to a quantum theory formalism . for the sake of example , we consider a specific weak amplification scheme @xcite , depicted in fig .
1 , which has been recently demonstrated experimentally @xcite . it aims at measuring very small temporal delays @xmath0 , or correspondingly tiny phase changes @xcite , with the help of optical pulses of much larger duration .
we consider this specific case because it contains the main ingredients of a typical wva scheme , explained below , and it allows to derive analytical expressions of all quantities involved , which facilitates the analysis of main results .
moreover , the scheme makes use of linear optics elements only and also works with large - bandwidth partially - coherent light @xcite .
in general , a wva scheme requires three main ingredients : a ) the consideration of two subsystems ( here two degrees of freedom : the polarisation and the spectrum of an optical pulse ) that are weakly coupled ( here we make use of a polarisation - dependent temporal delay that is introduced with the help of a michelson interferometer ) ; b ) the _ pre - selection _ of the input state of both subsystems ; and c ) the _ post - selection _ of the state in one of the subsystems ( the state of polarisation ) and the measurement of the state of the remaining subsystem ( the spectrum of the pulse ) . with appropriate _ pre- _ and _ post - selection _ of the polarisation of the output light , tiny changes of the temporal delay @xmath0 can cause anomalously large changes of its spectrum , rendering in principle detectable very small temporal delays . )
splits the input into two orthogonal linear polarisations that propagate along different arms of the interferometer .
an additional qwp is introduced in each arm to rotate the beam polarisation by @xmath1 to allow the recombination of both beams , delayed by a temporal delay @xmath0 , in a single beam by the same pbs .
after pbs@xmath2 , the output polarisation state is selected with a liquid crystal variable retarder ( lcvr ) followed by a second polarising beam splitter ( pbs@xmath3 ) .
the variable retarder is used to set the parameter @xmath4 experimentally .
finally , the spectrum of each output beam is measured using an optical spectrum analyzer ( osa ) .
( @xmath5,@xmath6 ) and ( @xmath7,@xmath8 ) correspond to two sets of orthogonal polarisations .
figure drawn by one of the authors ( luis - jose salazar serrano).,scaledwidth=70.0% ] let us be more specific about how all these ingredients are realized in the scheme depicted in fig .
1 . an input coherent laser beam ( @xmath9 photons ) shows circular polarisation , @xmath10 , and a gaussian shape with temporal width @xmath11 ( full - width - half maximum , @xmath12 ) .
the normalized temporal and spectral shapes of the pulse read @xmath13 the input beam is divided into the two arms of a michelson interferometer with the help of a polarising beam splitter ( pbs@xmath2 ) .
light beams with orthogonal polarisations traversing each arm of the interferometer are delayed @xmath14 and @xmath15 , respectively , which constitute the weak coupling between the two degrees of freedom .
after recombination of the two orthogonal signals in the same pbs@xmath2 , the combination of a liquid - crystal variable retarder ( lcvr ) and a second polarising beam splitter ( pbs@xmath3 ) performs the post - selection of the polarisation of the output state , projecting the incoming signal into the polarisation states @xmath16 $ ] and @xmath17 $ ] .
the amplitudes of the signals in the two output ports write ( not normalized ) @xmath18 \left\{1 + \exp \left [ i ( \omega_0+\omega ) \tau - i\gamma \right ] \right\ } \label{projections1 } \\ & & \phi_v(\tau)=\frac{\psi(\omega)}{2 } \exp \left [ i \left ( \omega_0 + \omega \right)\tau_0 \right ] \left\ { 1-\exp \left [ i ( \omega_0+\omega ) \tau
-i\gamma \right ] \right\ } , \label{projections2}\end{aligned}\ ] ] where @xmath19 .
( solid blue line ) and @xmath20 as ( dashed green line ) . in ( a ) the post - selection angle @xmath4 is @xmath21 , so as to fulfil the condition @xmath22 . in ( b )
the angle @xmath4 is @xmath23 .
( c ) shift of the centroid of the spectrum of the output pulse after projection into the polarisation state @xmath7 in pbs@xmath3 , as a function of the post - selection angle @xmath4 .
green solid line : @xmath24 as ; dotted red line : @xmath25 as , and dashed blue line : @xmath20 as .
label * i * corresponds to @xmath26 [ mode for @xmath20 as shown in ( b ) ] . label * ii * corresponds to @xmath27 , where the condition @xmath22 is fulfiled [ mode for @xmath20 shown in ( a ) ] .
it yields the minimum mode overlap between states with @xmath28 and @xmath29 .
data : @xmath30 m and @xmath31 fs.,scaledwidth=90.0% ] after the signal projection performed after pbs@xmath3 , the wva scheme distinguishes different states , corresponding to different values of the temporal delay @xmath0 , by measuring the spectrum of the outgoing signal in the selected output port .
the different spectra obtained for delays @xmath28 and @xmath20 as , for two different polarisation projections , are shown in figures 2 ( a ) and 2 ( b ) . to characterize different modes one can measure , for instance , the centroid of the spectrum .
2 ( c ) shows the centroid shift of the output signal for @xmath32 , which reads @xmath33 the differential power between both signals ( with @xmath28 and @xmath32 ) reads @xmath34\ ] ] when there is no polarisation - dependent time delay ( @xmath28 ) , the centroid of the spectrum of the output signal is the same than the centroid of the input laser beam , i.e. , there is no shift of the centroid ( @xmath35 ) .
however , the presence of a small @xmath0 can produce a large and measurable shift of the centroid of the spectrum of the signal .
detecting the presence ( @xmath29 ) or absence ( @xmath28 ) of a temporal delay between the two coherent orthogonally - polarised beams after recombination in pbs@xmath2 , but before traversing pbs@xmath3 , is equivalent to detecting which of the two quantum states , @xmath36 or @xmath37 is the output quantum state which describes the coherent pulse leaving pbs@xmath2 .
@xmath38 designates the corresponding polarisations . the spectral shape ( mode function )
@xmath39 writes @xmath40 , \label{modes_input}\ ] ] where @xmath41 is the central frequency of the laser pulse , @xmath42 is the angular frequency deviation from the the center frequency and @xmath43 is the spectral shape of the input coherent laser signal . the minimum probability of error that can be made when distinguishing between two quantum states is related to the trace distance between the states @xcite . for two pure state , @xmath44 and @xmath45 ,
the ( minimum ) probability of error is @xcite @xmath46 for @xmath47 , @xmath48 . on the contrary , to be successful in distinguishing two quantum states with low probability of error ( @xmath49 ) requires @xmath50 , i.e. , the two states should be close to orthogonal .
the coherent broadband states considered here can be generally described as single - mode quantum states where the mode is the corresponding spectral shape of the light pulse .
let us consider two single - mode coherent beams @xmath51 where @xmath52 and @xmath53 are the two modes @xmath54 and @xmath55 and @xmath56 are the mean number of photons in modes @xmath52 and @xmath53 , respectively .
the mode functions @xmath57 and @xmath58 are assumed to be normalized , i.e. , @xmath59 , reads @xmath60 where we introduce the mode overlap @xmath61 that reads @xmath62^{*}.\ ] ] in order to obtain eq .
( [ overlap1 ] ) we have made use of @xmath63^m |0 \rangle = n ! \rho^n \delta_{nm}$ ] . for @xmath64 ( coherent beams in the same mode but with possibly different mean photon numbers ) we recover the well - known formula for single - mode coherent beams @xcite : @xmath65 . making use of eqs .
( [ modes_input ] ) , ( [ overlap1 ] ) and ( [ overlap2 ] ) we obtain @xmath66 , \label{input_result}\end{aligned}\ ] ] where @xmath67 of the mode functions corresponding to the quantum states with @xmath28 and @xmath20 as , as a function of the post - selection angle @xmath4 ( solid blue line ) .
the insertion loss , given by @xmath68 is indicated by the dotted green line .
the minimum mode overlap , and maximum insertion loss , corresponds to the post - selection angle @xmath4 that fulfils the condition @xmath22 , which corresponds to @xmath27 .
data : @xmath69 m , @xmath31 fs.,scaledwidth=50.0% ] in the wva scheme considered here , the signal after pbs@xmath3 is projected into the orthogonal polarisation states @xmath7 and @xmath8 , and as a result the signals in both output ports are given by eqs .
( [ projections1 ] ) and ( [ projections2 ] ) . making use of eqs .
( [ projections1 ] ) , ( [ projections2 ] ) and ( [ overlap2 ] ) one obtains that the mode overlap ( for @xmath70 ) reads @xmath71}{2\left[1+\cos \gamma
\right]^{1/2 } \left[1+\gamma \cos ( \omega_0 \tau-\gamma ) \right]^{1/2}}.\ ] ] for @xmath28 , and therefore @xmath72 , we obtain @xmath64 .
3 shows the mode overlap of the signal in the corresponding output port for a delay of @xmath20 as .
the mode overlap has a minimum for @xmath22 , where the two mode functions becomes easily distinguishable , as shown in fig .
the effect of the polarisation projection , a key ingredient of the wva scheme , can be understood as a change of the mode overlap ( _ mode distinguishability _ ) between states with different delay @xmath0 . however , an enhanced mode distinguishability in this output port is accompanied by a corresponding increase of the insertion loss , as it can be seen in fig .
the insertion loss , @xmath73 $ ] , is the largest when the modes are close to orthogonal ( @xmath74 ) . both effects indeed compensate , as it should be , since wva implements unitary transformations , and the trace distance between quantum states is preserved under unitary transformations .
the quantum overlap between the states reads @xmath75,\end{aligned}\ ] ] so @xmath76 \label{output_result},\end{aligned}\ ] ] which is the same result [ see eq .
( [ input_result ] ) ] obtained for the signal after pbs@xmath2 , but before pbs@xmath3 .
we can also see the previous results from a slightly different perspective making use of the cramr - rao inequality @xcite .
the wva scheme considered throughout can be thought as a way of estimating the value of the single parameter @xmath0 with the help of a light pulse in a coherent state @xmath77 . since the quantum state is pure , the minimum variance that can show any unbiased estimation of the parameter @xmath0 , the cramr - rao inequality , reads @xmath78^{-1 } , \label{cramer1}\ ] ] making use of eq .
( [ state1 ] ) , one obtains that here the cramr - rao inequality reads @xcite @xmath79 where @xmath80 is the rms bandwidth in angular frequency of the pulse . in all cases of interest @xmath81 .
the cramr - rao inequality is a fundamental limit that set a bound to the minimum variance that any measurement can achieve .
it is unchanged by unitary transformations and only depends on the quantum state considered .
inspection of eqs .
( [ input_result ] ) and ( [ output_result ] ) seems to indicate that a measurement after projection in any basis , the core element of the weak amplification scheme , provides no fundamental metrological advantage .
notice that this result implies that the only relevant factor limiting the sensitivity of detection is the quantum nature of the light used ( a _ coherent state _ in our case ) . to obtain this result
, we are implicitly assuming that a ) we have full access to all relevant characteristics of the output signals ; and b ) detectors are ideal , and can detect any change , as small as it might be , if enough signal power is used . if this is the case , weak value amplification provides no enhancement of the sensitivity .
however , this can be far from truth in many realistic experimental situations . in the laboratory
, the quantum nature of light is an important factor , but not the only one , limiting the capacity to measure tiny changes of variables of interest . on the one hand ,
most of the times we detect only certain characteristic of the output signals , probably the most relevant , but this is still partial information about the quantum state . on the other hand , detectors are not ideal and noteworthy limitations to its performance can appear . to name a few , they might no longer work properly above a certain photon number input , electronics and signal processing of data can limit the resolution beyond what is allowed by the specific quantum nature of light , conditions in the laboratory can change randomly effectively reducing the sensitivity achievable in the experiment .
surely , all of these are _ technical _ rather than _ fundamental _ limitations , but in many situations the ultimate limit might be _ technical _ rather than _
fundamental_. in this scenario , we show below that weak value amplification can be a _
valuable _ and an _ easy _ option to overcome all of these technical limitations , as it has been demonstrated in numerous experiments . that leaves the interferometer .
the two points highlighted corresponds to @xmath82 , which yields @xmath83 , and @xmath84 , which yields @xmath85 .
( b ) number of photons ( @xmath86 ) after projection in the polarisation state @xmath87 $ ] , as a function of the angle @xmath4 .
the input number of photons is @xmath84 .
the dot corresponds to the point @xmath88 and @xmath89 .
pulse width : @xmath11=1 ps ; temporal delay : @xmath0= 1 as.,scaledwidth=90.0% ]
let us suppose that we have at hand light detectors that can not be used with more than @xmath90 photons .
any limitation on the detection time or the signal power would produce such limitation .
the technical advantages of using wva in this scenario has been previously pointed out @xcite .
here we make this apparent from a quantum estimation point of view , and quantify this advantage .
4(a ) shows the minimum probability of error as a function of the number of photons ( @xmath9 ) entering ( and leaving ) the interferometer . for @xmath91 ,
inspection of the figure shows that the probability of error is @xmath92 .
this is the best we can do with this experimental scheme and these particular detectors without resorting to weak value amplification .
however , if we project the output signal from the interferometer into a specific polarisation state , and increase the flux of photons , we can decrease the probability of error , without necessarily going to a regime of high depletion of the signal @xcite .
for instance , with @xmath89 , and a flux of photons of @xmath84 , so that after projection @xmath88 photons reach the detector , the probability of error is decreased to @xmath85 , effectively enhancing the sensitivity of the experimental scheme ( see fig .
the probability of error can be further decreased , also for other projections , at the expense of further increasing the input signal @xmath9 . in general ,
the minimum quantum overlap achievable between the states without any projection is @xmath93,\ ] ] while making use of projection in a weak value amplification scheme is @xmath94 .
\label{enhancement}\ ] ] eq .
( [ enhancement ] ) shows that when the number of photons that the detection scheme can handle is limited ( @xmath90 ) , projection into a particular polarisation state , at the expense of increasing the signal level , is advantageous . from a quantum estimation point of view , wva increases the minimum probability of error reachable , since the projection makes possible to use the maximum number of photons available ( @xmath90 ) with a corresponding enhanced mode overlap .
notice that the effect of using different polarisation projections can be beautifully understood as reshaping of the balance between signal level and mode overlap .
as second example , let us consider that specific experimental conditions makes hard , even impossible , to detect very similar modes , i.e. , with mode overlap @xmath95
. we can represent this by assuming that there is an _ effective _ mode overlap ( @xmath96 ) which takes into account all relevant experimental limitations of a specific set - up , given by @xmath97.\ ] ] fig .
5 shows an example where we assume that detected signals corresponding to @xmath98 can not be safely distinguished due to technical restrictions of the detection system . for @xmath98 , @xmath99 , so the detection system can not distinguish the states of interest even by increasing the level of the signal . on the contrary , for smaller values of @xmath61 , accessible making use of a weak amplification scheme , this limitation does not exist since the detection system can resolve this modes when enough signal is present .
the detection system can not distinguish the states of interest .
data : @xmath100 and @xmath101.,scaledwidth=50.0% ] up to now , we have used the concept of trace distance to look for the minimum probability of error achievable in _ any _ measurement when using a given quantum state . in doing that
, we only considered how the quantum state changes for different values of the variable to be measured , without any consideration of how this quantum state is going to be detected .
if we would like to include in the analysis additional characteristics of the detection scheme , one can use the concept of fisher information , that requires to consider the probability distribution of possible experimental outcomes for a given value of the variable of interest . in this approach ,
one chooses different probability distributions to describe formally _ characteristics _ of specific detection scheme @xcite .
let us assume that to estimate the value of the delay @xmath0 , we measure the shift of the centroid ( @xmath102 ) of the spectrum @xmath103 , given by eq .
( [ projections2 ] ) .
a particular detection scheme will obtain a set of results @xmath104 , @xmath105 for a given delay @xmath0 .
@xmath9 is the number of photons detected .
the fisher information @xmath106 provides a bound of @xmath107 for any unbiased estimator when the probability distribution @xmath108 of obtaining the set @xmath109 , for a given @xmath0 , is known .
if we assume that the probability distribution @xmath110 is gaussian , with mean value @xmath102 given by eq .
( [ centroid_shift ] ) and variance @xmath111 , determined by the errors inherent to the detection process , the fisher information reads @xcite @xmath112 ^ 2\ ] ] where @xmath113}{2\pi\ , \left ( 1+\gamma \cos \phi\right)^2}\ ] ] and @xmath114 .
for @xmath115 , i.e. , the angle of post - selection is @xmath116 , the fisher information is @xmath117 notice that @xmath118 corresponds to considering equal input and output polarization state , i.e. , no weak value amplification scheme . for @xmath119 , where the angle of post - selection is @xmath120 , we have @xmath121 @xmath122 corresponds to considering an output polarisation state orthogonal to the input polarisation state i.e. , when the effect of weak value amplification is most dramatic , as it can be easily observed in fig .
the fisher bound for @xmath123 is a factor @xmath124 larger than the bound for @xmath125 , so wva achieves enhancement of the fisher information .
this fisher information enhancement effect , which does not happen always , it has been observed for certain wva schemes @xcite .
there is no contradiction between the facts that the minimum probability of error , obtained by making use of the concept of trace distance , is not changed by wva , while at the same time there can be enhancement of the fisher information . by selecting a particular probability distribution to evaluate the fisher information , we include information about the detection scheme . in our case
, we estimate the value of @xmath0 by measuring the @xmath0-dependent shift of the centroid of the spectrum of the signal in one output port after pbs@xmath3 , which is only part of all the information available , given by the full signal in eqs .
( [ projections1 ] ) and ( [ projections2 ] ) .
we also assumed a gaussian probability distribution with a constant variance @xmath111 independent of @xmath0 .
the cramr - rao bound we have derived here depends on the full information available ( the quantum state ) before any particular detection . an unitary transformation , as wva is , does not modify the bound . on the contrary , the fisher information , by using a particular probability distribution to describe the possible outcomes in an particular experiment , selects certain aspects of the quantum state to be measured ( _ partial information _ ) , and this bound can change in a wva scheme , although the bound should be always above the cramr - rao bound . in this restrictive scenario ,
the use of certain polarization projections can be preferable .
the existence and nature of these different bounds might possibly explain certain confusion about the capabilities of wva , whether wva is considered to provide any metrological advantage or not .
on the one hand , if we consider the trace distance , or the quantum cramr - rao inequality , without any consideration about how the quantum states are detected , post - selection inherent in wva does not lower the minimum probability of error achievable , so from this point of view wva offers no metrological advantage . on the other hand , in certain scenarios , the fisher information , when it takes into account _ information about the detection scheme _ ,
can be enhanced due to post - selection . in this sense
, one can think of wva as an advantageous way to optimize a particular detection scheme .
wva schemes makes use of linear optics unitary transformations . therefore , if the only limitations in a measurement are due to the quantum nature ( _ intrinsic statistics _ ) of the light , for instance , the presence of shot noise in the case of coherent beams , wva does not offer any advantage regarding any decrease of the minimum probability of error achievable .
this is shown by making use of the trace distance between quantum states or the cramr - rao inequality , which set sensitivity bounds that are independent of any particular post - selection .
however , notice that this implicitly assume that full information about the quantum states used can be made available , and detectors are ideal , so they can detect any change of the variable of interest , as small as it might be , provided there is enough signal power . nevertheless , these assumptions are in many situations of interest far from true . these limitations , sometimes refereed as _ technical noise _ , even though not fundamental ( one can always imagine using a better detector or a different detection scheme ) are nonetheless important , since they limit the accuracy of specific detection systems at hand . in these scenarios ,
the importance of weak value amplification is that by decreasing the mode overlap associated with the states to be measured and possibly increasing the intensity of the signal , the weak value amplification scheme allows , in principle , to distinguish them with lower probability of error .
we have explored some of these scenarios from an quantum estimation theory point of view .
for instance , we have seen that when the number of photons usable in the measurement is limited , the minimum probability of error achievable can be effectively decreased with weak value amplification .
we have also analyzed how weak value amplification can differentiate between _ in practice_-indistinguishable states by decreasing the mode overlap between its corresponding mode functions .
finally we have discussed how the confusion about the usefulness of weak value amplification can possibly derive from considering different bounds related to how much sensitivity can , in principle , be achieved when estimating a certain variable of interest .
one might possibly say that the advantages of wva _ have nothing to do with fundamental limits and should not be viewed as addressing fundamental questions of quantum mechanics _
@xcite . however , _ from a practical rather than fundamental point of view _
, the use of wva can be advantageous in experiments where sensitivity is limited by experimental ( technical ) , rather than fundamental , uncertainties . in any case
, if a certain measurement is _ optimum _ depends on its capability to effectively reach any bound that might exist .
* references * 19 aharonov , y. , albert , d. z. & vaidman , l. how the result of a measurement of a component of the spin of a spin-1/2 particle can turn out to be 100 .
_ * 60 * , 13511354 ( 1988 ) .
hosten , o. & kwiat , p. observation of the spin hall effect of light via weak measurements , _ science _ * 319 * , 787790 ( 2008 ) .
zhou , x. , zhou , ling , x. , luo h. , & wen , s. identifying graphene layers via spin hall effect of light .
* 101 * , 251602 ( 2012 ) .
ben dixon p. , starling , d. j. , jordan , a. n. , & howell , j. c. ultrasensitive beam deflection measurement via interferometric weak value amplification .
* 102 * , 173601 ( 2009 ) .
pfeifer , m. , & fischer , p. weak value amplified optical activity measurements .
express _ * 19 * , 1650816517 ( 2011 ) . howell , j. c. , starling , d. j. , dixon , p. b. , vudyasetu , k. p. & jordan , a. n. precision frequency measurements with interferometric weak values .
a _ * 82 * , 063822 ( 2010 ) .
egan , p. & stone , j. a. weak - value thermostat with 0.2 mk precision .
lett . _ * 37 * , 49914993 ( 2012 ) .
xu , x. y. et al .
phase estimation with weak measurement using a white light source
lett . _ * 111 * , 033604 ( 2013 ) .
dressel , j. , malik , m. , miatto , f. m. , jordan , a. n. & boyd r. w. colloquium : understanding quantum weak values : basics and applications .
86 * , 307316 ( 2014 ) .
jordan , a. n. , martez - rincn , j. & howell , j. c. technical advantages for weak - value amplification : when less is more .
x * 4 * , 011031 ( 2014 ) .
knee , g. c. , & gauger , e. m. when amplification with weak values fails to suppress technical noise .
x * 4 * , 011032 ( 2014 ) .
ferrie , c. & combes , j. weak value amplification is suboptimal for estimation and detection . _
lett . _ * 112 * , 040406 ( 2014 ) .
vaidman , l. comment on _ weak value amplification is sub - optimal for estimation and detection .
lett . _ * 111 * , 033604 ( 2013)_. arxiv:1402.0199v1 [ quant - ph ] ( 2014 ) .
helstrom , c. w. _ quantum detection and estimation theory _ , academic press inc .
nielse , m. a. & chuang i. l. _ quantum computation and quantum information _ , cambridge university press , 2000 .
zhang , l. , datta , a. & walsmely , i. a. precision metrology using weak measurements .
lett . _ * 114 * , 210801 ( 2015 ) torres , j. p. , puentes , g. , hermosa , n. & salazar - serrano , l. j. weak interference in the high - signal regime . _
express _ * 20 * , 18869 - 18875 ( 2012 ) .
duck , i. m. , stevenson , p. m. & sudarhshan , e. c. g. the sense in which a weak measurement of a spin-1/2 particles s spin component yields a value of 100 .
d * 40 * , 21122117 ( 1989 ) .
howell , j. c. , starling , d. j. , dixon , p. b. , vudyasetu , k. p. & jordan , a. n. interferometric weak value deflections : quantum and classical treatments .
rev . a _ * 81 * , 033813 ( 2010 )
. ritchie , n. w. , story , j. g. & hulet , r. g. realization of a measuremernt of a weak value .
lett . _ * 66 * 11071110 ( 1991 ) .
brunner , n. & simon , c. measuring small longitudinal phase shifts : weak measurements of standard interferometry .
_ * 105 * 010405 ( 2010 ) .
salazar - serrano , l. j. , janner , d. , brunner , n. , pruneri , v. & torres , j. p. measurement of sub - pulse - width temporal delays via spectral interference induced by weak value amplification .
a _ * 89 * 012126 ( 2014 ) .
strubi , g. & bruder , c. measuring ultrasmall time delays of light by joint weak measurements .
lett . _ * 110 * 083605 ( 2012 ) .
li , c .- f .
ultrasensitive phase estimation with white light .
rev . a _ * 83 * , 044102 ( 2011 ) . fuchs , c. a. , & van de graaf , j. cryptographic distinguishability measures for quantum mechanical states .
_ ieee t. inform . theory _
* 45 * , 12161227 ( 1999 ) .
englert , b .-
fringe visibility and which - way information : an inequality , _ phys .
lett . _ * 77 * , 21542157 ( 1996 ) .
ou , z. y. complementarity and fundamental limit in precision phase measurement .
_ phys . rev .
_ * 77 * , 23522355 ( 1996 ) .
glauber , r. j. coherent and incoherent states of the radiation field .
_ * 131 * , 27662788 ( 1966 ) .
let us define @xmath126 .
for a coherent product state of the form @xmath127 , where the index @xmath128 refers to different frequency modes , one obtains that @xmath129 , where @xmath130 .
if @xmath131 , where @xmath132 is the mean number of photons in frequency mode @xmath128 and only @xmath133 depends on the parameter @xmath0 as @xmath134 , one obtains that @xmath135 , where @xmath136 is the creation operator of the corresponding frequency mode . the unknown parameter of interest has value @xmath0 .
after repeated measurements to estimate its value , we obtain a distribution of outcomes @xmath137 which can be characterized by a probability distribution @xmath138 that depends on the value of @xmath0 .
the fisher information reads @xmath139 .
the variance of any unbiased estimator that makes use of the ensemble @xmath137 is bounded from below by @xmath140 . when the fisher function can be written as @xmath141 $ ] , where @xmath142 is the variable that we measure , the fisher information can be written as @xmath143 .
viza , g. i. et al .
weak - values technique for velocity measurements .
_ * 38 * , 2949 - 2952 ( 2013 ) .
combes , j. & ferrie , c. & zhang , j. , and carlton m. caves , c. m. quantum limits on postselected , probabilistic quantum metrology , _ phys .
a _ * 89 * , 052117 ( 2014 ) .
weak value amplification scheme aimed at detecting extremely small temporal delays .
the input pulse polarisation state is selected to be left - circular by using a polariser , a quarter - wave plate ( qwp ) and a half - wave plate ( hwp ) .
a first polarising beam splitter ( pbs@xmath2 ) splits the input into two orthogonal linear polarisations that propagate along different arms of the interferometer .
an additional qwp is introduced in each arm to rotate the beam polarisation by @xmath1 to allow the recombination of both beams , delayed by a temporal delay @xmath0 , in a single beam by the same pbs .
after pbs@xmath2 , the output polarisation state is selected with a liquid crystal variable retarder ( lcvr ) followed by a second polarising beam splitter ( pbs@xmath3 ) .
the variable retarder is used to set the parameter @xmath4 experimentally .
finally , the spectrum of each output beam is measured using an optical spectrum analyzer ( osa ) .
( @xmath5,@xmath6 ) and ( @xmath7,@xmath8 ) correspond to two sets of orthogonal polarisations .
spectrum measured at the output .
( a ) and ( b ) : spectral shape of the mode functions for @xmath28 ( solid blue line ) and @xmath20 as ( dashed green line ) . in ( a ) the post - selection angle
@xmath4 is @xmath21 , so as to fulfil the condition @xmath22 . in ( b )
the angle @xmath4 is @xmath23 .
( c ) shift of the centroid of the spectrum of the output pulse after projection into the polarisation state @xmath7 in pbs@xmath3 , as a function of the post - selection angle @xmath4 .
green solid line : @xmath24 as ; dotted red line : @xmath25 as , and dashed blue line : @xmath20 as .
label * i * corresponds to @xmath26 [ mode for @xmath20 as shown in ( b ) ] .
label * ii * corresponds to @xmath27 , where the condition @xmath22 is fulfiled [ mode for @xmath20 shown in ( a ) ] .
it yields the minimum mode overlap between states with @xmath28 and @xmath29 .
data : @xmath30 m and @xmath31 fs . mode overlap and insertion loss as a function of the post - selection angle .
mode overlap @xmath61 of the mode functions corresponding to the quantum states with @xmath28 and @xmath20 as , as a function of the post - selection angle @xmath4 ( solid blue line ) . the insertion loss , given by @xmath68
is indicated by the dotted green line .
the minimum mode overlap , and maximum insertion loss , corresponds to the post - selection angle @xmath4 that fulfils the condition @xmath22 , which corresponds to @xmath27 .
data : @xmath69 m , @xmath31 fs . reduction of the probability of error using a weak value amplification scheme .
( a ) minimum probability of error as a function of the photon number @xmath9 that leaves the interferometer .
the two points highlighted corresponds to @xmath82 , which yields @xmath83 , and @xmath84 , which yields @xmath85 .
( b ) number of photons ( @xmath86 ) after projection in the polarisation state @xmath87 $ ] , as a function of the angle @xmath4 .
the input number of photons is @xmath84 .
the dot corresponds to the point @xmath88 and @xmath89 .
pulse width : @xmath11=1 ps ; temporal delay : @xmath0= 1 as . | weak value amplification ( wva ) is a concept that has been extensively used in a myriad of applications with the aim of rendering measurable tiny changes of a variable of interest . in spite of this
, there is still an on - going debate about its _ true _ nature and whether is really needed for achieving high sensitivity .
here we aim at solving the puzzle , using some basic concepts from quantum estimation theory , highlighting what the use of the wva concept can offer and what it can not .
while wva can not be used to go beyond some fundamental sensitivity limits that arise from considering the full nature of the quantum states , wva can notwithstanding enhance the sensitivity of _ real _ detection schemes that are limited by many other things apart from the quantum nature of the states involved , i.e. _ technical noise_. importantly , it can do that in a straightforward and easily accessible manner . |
the atom interferometry used in this proposal is similar to @xcite and uses many of the techniques in @xcite .
light pulse atom interferometry uses two counter - propagating lasers that couple hyperfine degenerate ground states of alkali ( or alkali - like ) atoms through a near - resonance raman transition . while the lasers are on , the system undergoes rabi oscillations between two states having a relative momentum @xmath12 . by performing a @xmath13 series of raman pulses ,
the atom s wave packet is split into a slow and fast component , then after an interrogation time , @xmath14 , the states are reversed and the wave packet is brought back together for the final beam splitter that interferes the two halves of the wave packet . the maximal spatial separation of the wave packets in the interferometer is @xmath15 .
there often is an initial velocity , @xmath16 , to the atom s wave packets that is used to doppler - select the desired atomic transitions .
an initial velocity can also arise from the thermal velocities .
atom interferometers are always run in pairs with the same lasers driving both interferometers in order to remove laser phase noise .
the paired interferometers also reduce other common mode backgrounds such as the sagnac effect .
the distance between these interferometers can be large and a benchmark value of several meters will be used .
finally , interferometers are run with @xmath17 atoms simultaneously in a bunch and the full experiment is performed @xmath18 times .
@xmath17 is determined by the rate for cooling atoms and @xmath19 is currently possible .
the interrogation time for the experiment will be @xmath20 , so a benchmark values of @xmath21 is reasonable . in several days
@xmath22 trials can be run and gives a shot - noise phase sensitivity of @xmath23 .
atomic fountains are frequently used in ai and start with an ensemble of evaporatively cooled atoms ( @xmath24 ) in a magneto - optical trap ( mot ) that localizes that atoms to @xmath25 .
the atoms are then launched with a velocity , @xmath26 , vertically through a series of multiple bragg or raman transitions .
the atoms subsequently follow ballistic motion for an interrogation time , @xmath27 , defining @xmath26 at the initial @xmath28 pules and performing the @xmath29 pulse occurs at the apex of the trajectory . in atomic fountains ,
the atoms are instantaneously in free fall and decoupled from their environment except during periods when they are coupled to the lasers which last .
the atoms are more isolated than meso- or macroscopic measurements that are vibrationally coupled to the environment and essentially removes the `` chopping '' used to isolate torsional pendulums or cantilevers from the environment .
configuring the interferometer as a gyroscope , with @xmath26 perpendicular to @xmath30 and @xmath16 , allows the atom s ballistic trajectory to be parallel to a planar face of the mass .
this maximizes the time that the atoms spend close to the source mass , thus making a signal from a new short distance force as large as possible .
the interferometer is sensitive to the potential as a function of distance away from the surface of the source mass if the recoil velocity is perpendicular to this face . by mapping out the potential
, it is possible to look for new contributions beyond the standard model .
this configuration minimizes the earth s gravity ; however , the phase from the corriolis force ( sagnac effect ) is maximized and controlling this is an important background discussed below . and
@xmath31 from the nearer of the paired interferometers .
the plane in gold represents a thin casimir shield fitted with retroreflectors that allow optical access for the right moving laser.,width=288 ] the relative phase between the two paths arises from several different contributions and not many exact results are known .
there is a semi - classical , perturbative method for computing the phase differences and at lowest order in the potential , @xmath32 , speed of light , @xmath33 , and the width of the wave packet , @xmath34 , the general result is that the phase difference is integral of the perturbing potential over the unperturbed paths .
there are several phase difference results that will be useful in deriving the sensitivity .
the first is if the potential only depends on the position in the direction of @xmath30 and where @xmath16 is not important @xmath35 where the initial position of the atom is taken to be @xmath36 . in the second half
the potential has been taylor expanded , if applicable .
the first term is proportional to the acceleration , and the second term is referred to as the recoil phase because it vanishes as @xmath37 becomes large , keeping @xmath38 fixed .
finally , if @xmath16 is important , the above expression becomes @xmath39 where @xmath40 , @xmath41 , and @xmath42 .
if the potential depends on the distance in the @xmath26 direction , no closed form is possible , but simple expressions can be obtained if @xmath43 is taylor expandable .
newton s constant is not known to a precision better than @xmath44 , so it is necessary to perform a series of measurements to remove absolute sensitivity to @xmath45 . the most straightforward manner to remove the uncertainty in @xmath45 is to test the @xmath46 behavior of gravity by having a moveable source mass .
the source mass will be taken to be a cylinder of radius , @xmath47 , and width , @xmath48 , and density , @xmath49 , with the circular face of the cylinder forming a vertical plane .
this geometry is motivated by calculational simplicity and none of the results depend sensitively upon the geometry
. the height of the atomic fountains , @xmath50 , will take place near the center of the cylinder s face .
the newtonian potential near the center of the cylinder can be calculated in the far field , @xmath51 , and near field , @xmath52 , limits near the center of the cylinder . in the far field limit @xmath53 where @xmath54 is
the distance to cylinder s face and @xmath55 is the distance from the cylinder s center .
correction to the potential is due to the cylindrical symmetry and will not be important for the results . ] in the near field limit the newtonian potential is given by @xmath56 where @xmath57 is an unmeasurable function , independent of the distance from the source .
there are several limits necessary for the yukawa potential .
the first is @xmath58 where the potential from cylinder is @xmath59 for @xmath60 , the yukawa potential looks newtonian plus a correction arising from taylor expanding the exponential . in the far field limit ( @xmath61 ) the yukawa potential is @xmath62 notice the absence of an @xmath63 term because it is always an unmeasurable constant . in the near field limit ( @xmath64 ) , the yukawa potential becomes @xmath65 the physical size of the experiment ultimately limit the sensitivity .
the size of the source mass sets the @xmath4 with maximum sensitivity and the benchmark value used is @xmath66 and @xmath67 .
the other relevant physical constraint is how near the source mass can get to the interferometers , @xmath68 .
the distance that the source mass can be moved from the interferometers does not limit the sensitivity so long as @xmath69 .
the strategy to distinguish newtonian gravity from a new yukawa potential is to perform a near and a far measurement of the phase .
the near measurement fixes what the newtonian prediction is for the far measurement .
if the inclusion of a yukawa potential with strength @xmath3 and range @xmath4 causes a difference between the far prediction and the far measurement greater than the shot noise limit , then this yukawa potential is visible assuming that no backgrounds or uncertainties are larger .
the scaling of the limits depend on @xmath4 and there two cases to be considered .
if , @xmath70 , it is possible to get within the @xmath46 behavior of the yukawa potential and move outside its range ; furthermore , the radius of the cylinder is not determining the sensitivity . in the second case @xmath71 , it is possible to get within the @xmath46 and outside the range of the potential , but the size of the source mass is determining the sensitivity . for large @xmath4 , it may seem beneficial to move the source mass further from the source to make the difference between a yukawa potential and the newtonian potential more pronounced , growing as @xmath72 ; however , the size of the newtonian phase shift is falling as @xmath73 meaning that there is no parametric gain for moving the source mass a distance greater than @xmath47 from the interferometers for any value of @xmath4 .
the biggest challenge of this experiment is the strategy of using a near measurement to extrapolate to a far prediction .
the challenge is knowing both the source mass position and orientation precisely enough to make a @xmath74 or better prediction for the far newtonian prediction . examining the subleading terms in the newtonian potentials in ( [ eq :
newton 1 ] ) and ( [ eq : newton 2 ] ) , if there is an uncertainty in the position of the cylinder of @xmath75 , then there is an uncertainty in the newtonian potential of @xmath76 or @xmath77 , respectively . by having @xmath78 reduces the uncertainty in the newtonian prediction from uncertainty in the initial height of the atom .
it is challenging but achievable to know the position of the block to @xmath79 to be sensitive to @xmath80 ; however , future improvements in sensitivity will not be limited by this . a better solution to knowing the position and orientation of the block is to use multiple interferometers situated near @xmath81 to actively measure the position and orientation of the source .
since the edge effects become @xmath82 near the edges , it is possible to use these additional measurements to locate the block and then use a central interferometer to use the near measurement to make the far prediction . this strategy will require six additional interferometers to over - determine the source s solid body coordinates .
the remaining challenge is to keep these seven interferometers locked in place as the source mass moves , but this should be possible to a much greater accuracy .
there is a residual uncertainty coming from `` jitter '' in the location of the mots .
this motion is of the order of @xmath83 ; however , it is stochastic with respect to the number of bunches of atoms , @xmath18 , and therefore the uncertainty is @xmath84 and only limiting after several rounds of improvements . the sagnac effect
, the phase induced by the corriolis force is large , @xmath85 , where @xmath86 is the angular velocity of the earth .
the use of paired interferometers described above cancels the leading order sagnac effect .
the novel method of actively reducing the sagnac effect is to rotate the lasers to compensate for the rotation of the earth .
it is possible to rotate lasers with nanoradian precision which reduces the sagnac effect by a factor of @xmath44 .
the dominant way that the sagnac effect contaminates the signal is from variation in relative @xmath26 of the two interferometers which is stochastic in the number of bunches , reducing the size by @xmath87 .
the final reduction in sagnac contamination must come from vibration isolation and requires @xmath88 .
the casimir potential between an atom and a conducting plate is given by @xmath89 , with @xmath90 being the polarizability of the atom and is @xmath91 for cs@xcite .
the phase difference arising from the casimir potential arises from ( [ eq : pot phase ] ) or ( [ eq : phase vi ] ) rather than ( [ eq : acc phase ] ) because of its rapid fall off .
these reduce the size of the casimir force relative to a macroscopic measurement device and it is not important for distances , ( and consequently @xmath4 ) greater than @xmath92 , but becomes important at shorter distances . in order to gain to reduce any residual phases , having a thin casimir shield ( @xmath93 in width ) that isolates the atom from the source mass position is helpful . while the the casimir potential from the shield will still be measurable , it will not be correlated with the mass position and therefore not directly a systematic effect . the primary way the casimir potential enters as a background is by having the source mass
deflect the shield and additionally through the `` jitter '' of the mots .
the deflection of the shield is sufficiently small so long as the source mass is @xmath94 from the shield .
the jitter gives a phase uncertainty of @xmath95 for @xmath96 and is sufficiently small .. ] in addition to isolating the atoms , it will be necessary to attach retroreflectors on the shield to provide optical access for the outward propagating lasers necessary to drive the raman transitions .
with the results of the previous section , the sensitivity to @xmath3 and @xmath4 can be computed . using the previously stated benchmarks for cs the phase sensitivity of @xmath97 from @xmath98 , @xmath99 , @xmath100 , @xmath101 , @xmath102 for pb and @xmath103 , the sensitivity curve is plotted in fig . [
fig : sensitivity ] relative in blue to current experimental limits shown black .
the peak sensitivity is @xmath104 and occurs for @xmath105 which is determined by the size of the source mass . for @xmath106 ,
current ai is more sensitive than existing experiments .
additionally , from @xmath107 current ai techniques matches the experimental limits .
this motivates considering future experimental improvements to if sensitivity is possible down to @xmath108 . in the near future it will be possible to increase the magnitude of the momentum transferred to the atoms in the interferometer using a technique called large momentum transfer ( lmt ) .
lmt uses repeated raman or bragg transitions during the laser pulses that impart on the atoms up to 100 times more momentum as a single raman transition .
lmt significantly impacts sensitivity for larger @xmath4 , increasing sensitivity by a factor of 100 ; however , for shorter wavelengths , where @xmath109 , there is no significant gain to sensitivity because the atoms immediately leave the region being sourced by yukawa potential .
, from either a need to doppler - select atomic transitions or thermal motion , there is a small gain at short distances proportional to @xmath110 shown in ( [ eq : phase vi ] ) . ]
the sensitivity is shown in fig .
[ fig : sensitivity ] in green
. on the short distance front , increases in sensitivity must come through increases in phase sensitivity .
one possibility is to increase the number of atoms being used in the experiment , which is limited by the length of the experiment and the rate for cooling atoms . in the future ,
a more effective way to increase sensitivity would be to use wave packets of entangled atoms where it is possible to have the sensitivity scale as @xmath111 ( in contrast to @xmath112 ) which is known as heisenberg limited statistics ( hs ) .
hs potentially allows significant gains in sensitivity and the sensitivity curve is shown for a phase sensitivity of @xmath113 in red in fig .
[ fig : sensitivity ] . in yellow ,
the sensitivity is shown for combining a lmt factor 100 and @xmath113 ; however , at this level the stochastic uncertainty in the newtonian prediction starts to become limiting .
there are some caveats that may limit the sensitivity at shorter distances .
the first is that all of the phase results are computed in the semi - classical limit ; however , when @xmath114 this approximation breaks down and a more complete quantum mechanical treatment is necessary .
additionally , the interaction between the casimir shield and atoms must be considered in much more detail , particularly for the case of hs where preventing the decoherence of the wave packet is necessary .
this letter has demonstrated that atom interferometry has the potential to improve the sensitivity to new forces in the @xmath115 to @xmath116 range using current technology or technologies that will be available in the near future .
additionally , atom interferometry can be useful in testing other types of forces other than the standard @xmath117 yukawa potential such as spin , velocity and composition dependent forces .
jw would like to thank m. kasevich , p. graham and s. dimopoulos for collaboration during early stages of this work and also g. biederman , b. dobrescu , and j. hogan for illuminating conversations .
jw is supported by the doe under contract de - ac03 - 76sf00515 .
jw is supported by the doe outstanding junior investigator award .
e. g. adelberger , b. r. heckel and a. e. nelson , ann .
sci . * 53 * , 77 ( 2003 ) [ arxiv : hep - ph/0307284 ] .
n. arkani - hamed , s. dimopoulos and g. r. dvali , phys .
d * 59 * , 086004 ( 1999 ) [ arxiv : hep - ph/9807344 ] .
i. antoniadis , n. arkani - hamed , s. dimopoulos and g. r. dvali , phys .
b * 436 * , 257 ( 1998 ) [ arxiv : hep - ph/9804398 ] .
n. arkani - hamed , s. dimopoulos and g. r. dvali , phys .
b * 429 * , 263 ( 1998 ) [ arxiv : hep - ph/9803315 ] .
r. d. peccei and h. r. quinn , phys .
* 38 * , 1440 ( 1977 ) .
f. wilczek , phys .
* 40 * , 279 ( 1978 ) .
m. dine , w. fischler and m. srednicki , phys .
b * 104 * , 199 ( 1981 ) .
m. a. shifman , a. i. vainshtein and v. i. zakharov , nucl .
b * 166 * , 493 ( 1980 ) .
j. e. kim , phys .
* 43 * , 103 ( 1979 ) .
j. e. moody and f. wilczek , phys .
d * 30 * , 130 ( 1984 ) .
r. barbieri , a. romanino and a. strumia , phys .
b * 387 * , 310 ( 1996 ) [ arxiv : hep - ph/9605368 ] .
m. pospelov , phys .
d * 58 * , 097703 ( 1998 ) [ arxiv : hep - ph/9707431 ] .
b. a. dobrescu and i. mocioiu , jhep * 0611 * , 005 ( 2006 ) [ arxiv : hep - ph/0605342 ] .
s. dimopoulos and a. a. geraci , phys .
d * 68 * , 124021 ( 2003 ) [ arxiv : hep - ph/0306168 ] .
q. g. bailey and v. a. kostelecky , phys .
d * 74 * , 045001 ( 2006 ) [ arxiv : gr - qc/0603030 ] .
g. biedermann , `` gravity tests , differential accelerometry and interleaved clocks with cold atom interferometers , '' ( 2008 ) .
k. y. chung , s. w. chiow , s. herrmann , s. chu and h. muller , phys .
d * 80 * , 016002 ( 2009 ) [ arxiv:0905.1929 [ gr - qc ] ] .
h. muller , s. w. chiow , s. herrmann , s. chu and k. y. chung , phys .
lett . * 100 * , 031101 ( 2008 ) [ arxiv:0710.3768 [ gr - qc ] ] .
s. dimopoulos , p. w. graham , j. m. hogan and m. a. kasevich , arxiv:0802.4098 [ hep - ph ] .
s. dimopoulos , p. w. graham , j. m. hogan and m. a. kasevich , phys .
lett . * 98 * , 111102 ( 2007 ) [ arxiv : gr - qc/0610047 ] . a. arvanitaki , s. dimopoulos , a. a. geraci , j. hogan and m. kasevich , phys .
lett . *
100 * , 120407 ( 2008 ) [ arxiv:0711.4636 [ hep - ph ] ] .
s. dimopoulos , p. w. graham , j. m. hogan , m. a. kasevich and s. rajendran , phys .
d * 78 * , 122002 ( 2008 ) [ arxiv:0806.2125 [ gr - qc ] ] .
s. dimopoulos , p. w. graham , j. m. hogan , m. a. kasevich and s. rajendran , phys .
b * 678 * , 37 ( 2009 ) [ arxiv:0712.1250 [ gr - qc ] ] .
berman , ed .
_ atom interferometry _
( new york : acad .
pr . , 1997 ) . | atom interferometry is a rapidly advancing field and this letter proposes an experiment based on existing technology that can search for new short distance forces . with current technology
it is possible to improve the sensitivity by up to a factor of @xmath0 and near - future advances will be able to rewrite the limits for forces with ranges from 100 @xmath1 m to 1 km .
new short distance forces are a frequent prediction of theories beyond the standard model and the search for these new forces is a promising channel for discovering new physics . over the past 15 years
there has been rapid advances in light pulse atom interferometry ( ai ) and in a wide variety of settings , ai is the most sensitive measurement .
this letter will explore the sensitivity of ai to new forces .
ai holds great promise in improving currently sensitivity over a wide range of distances from roughly to .
new forces can couple to matter in many different ways ; however , there is a benchmark parameterization that is frequently applicable to new forces where the potential between two particles is proportional to the mass of the particles @xmath2 where @xmath3 is a dimensionless number that characterizes the new force s strength relative to gravity and @xmath4 is the compton wavelength of the particle being exchanged . the coupling , @xmath3 , could be composition , spin or velocity dependent or have a power - law fall off rather than an exponential / yukawa behavior ; however , this parameterization is a standard benchmark and will be used in this letter .
theories predict a wide range of @xmath4 and @xmath3 .
some theories give @xmath5 such as gauge mediated supersymmetry theories that have moduli mediated forces @xcite , large extra dimensions @xcite or theories that have gravity shut off at the scale of the cosmological constant @xcite .
alternatively , many theories also predict @xmath6@xcite .
the most reknowned of these theories are peccei - quinn axions can mediate forces with @xmath7 @xcite .
thus , while it is important to continue the search for @xmath8 to shorter distance forces , searching for sub -gravitational strength forces is also an important frontier to continue pursuing .
finally , there are forces that are not yukawa forces of ( [ eq : yukawa ] ) and may intrinsically be stronger than gravity , but at long distances may show up as sub - gravitational strength forces @xcite .
see @xcite for other applications of atom interferometry to modifications of gravity .
atom interferometry uses cold atoms that have their quantum mechanical wave packets spatially split in two and recombined .
the final interference pattern measures the phase difference between the two paths .
the experiment described in this letter uses a source mass to create a potential that causes a relative phase between the two paths . by subtracting off the newtonian potential and other backgrounds ,
a new yukawa potential is visible .
the ai experiment in this letter is effective at probing new forces in the to range with sensitivities down to @xmath9 with already proven techniques in contrast to current experimental limits that have sensitivities of @xmath10@xcite .
future improvements can increase the range of sensitivity to to and with sensitivities down to @xmath11 . |
dark matter is a very strange form of matter . it does not interact with known objects because of the lack of electromagnetic interaction .
however , it has a mass and we know that it exists.the main reason for this view is from gravitational interactions on the visible matter which holds galaxies and galaxies of clusters together .
the idea of the existence of a new form of matter escaping electromagnetic detections , traces back at least to 1915 .
the name `` dark matter '' was firstly used by the jacobus kapteyn in 1922 in his studies of the motions of stars in our galaxy @xcite .
he found that there is no need of dark matter within and around the solar system . in 1923 , jan oort arrived to an opposite conclusion that there should be twice as much dark matter as visible matter in the solar vicinity .
this is the first claim of evidence for dark matter .
however , since later observations have disproved this early claim , in 1933 the discovery of dark matter is usually credited to fritz zwicky who made the first correct claim about the existence of dark matter.he concluded that the velocities of the galaxies in the coma cluster @xcite are much larger than the expected one .
zwicky found that orbital velocities are almost a factor of ten larger than expected from the total mass of the cluster @xcite .
zwicky concluded that , in order to hold galaxies together in the cluster , the cluster should contain huge amounts of some dark ( invisible ) matter .
the galaxy rotation curve v , where r marks the distance from the galaxy s center .
@xmath1 where m(r ) is the mass inside radius r and @xmath2 .
after rearranged the above equation reads @xmath3 if the mass is only inside the radius , the velocity should decrease when there is large value of radius as @xmath4 , where @xmath5 .
however , for values of r bigger than the size of the galactic bulge , the velocity of stars remains approximately constant@xcite .
it means that there is a some unknown energy there such that @xmath6 ] for an orbit around a compact central mass , such as planets in the solar system , we get kepler s third law which is given at eqn.(1.3 ) . as an example if the star is placed in the center of a galaxy , that is the mass density of a galaxy decreases as a power law i.e
@xmath7 the mass inside radius r is @xmath8 thus the rotation velocity is @xmath9 near the center of the galaxy , observed rotation curves of a galaxy v(r ) increase with r from eqn.(1.3 ) , however then suddenly becomes @xmath10 constant . from the eqn.(1.5 ) , the density profile is @xmath11 on the other side , the density of stars falls faster when goes to the edges of the galaxy .
the total mass from stars , gas and dust clouds , is very very small so that they are not able to provide the rotational speed at large distances .
this means that there is another invisible , massive matter
. this matter would be only dominant in the outer parts .
the dark component forms a dark halo which holds galaxies and galaxies of clusters together . ] at galactic scales , there is also additional evidence for dark matter which is based on the mass modeling of the detailed rotation curves @xcite .
an evidence of dark matter from cluster of galaxies was first noticed , as mentioned already , by @xcite in 1933 . on the cluster scales ,
the @xmath12 is todays mass density including baryonic mass and dark matter @xcite @xmath13 ] . as known that matter relic density component is bounded by the observation of the large scale structure.as reported by sloan digital sky survey ( sdss ) @xmath14 for @xmath15 @xcite .
the mass density of baryons and the mass density of matter in the universe by using wmap data are found to be @xcite + @xmath16 @xmath17 ] from wmaps data and large scale structe observationss data for relic abundance are consistent . in physical cosmology ,
big bang nucleosynthesis provides us also a constraint@xcite the difference between @xmath18 and @xmath19 is the clue of the existence of the non - baryonic dark matter .
there are many experiments which support the existence of dark matter such as cmb anisotropy , galaxy surveys , etc .
so we think that there is dark matter .
unfortunately , the nature of dark matter is now an open problem . so
this is a very hot topic in physics . furthermore , there are some additional constraints on the amount of dark matter in the universe .
we know very well from the concordance cosmology measurements(it s basically a combination of looking at the microwave background fluctuations , the baryon acoustic oscillations , and the supernova brightness versus redshift relation)what are the relative fractions of dark energy and mass in the universe .
we also have pretty good constraints from primordial abundances of the elements on what the relative fractions are for baryonic and non - baryonic dark matter because if there were too high a density of protons in the early universe , then there would have been more fusion in the early universe@xcite .
macho(massive compact halo object ) ; planet - like objects , plasma and other compact objects are the candidates for bdm
. other examples of this class of dark matter candidates include primordial black holes created during the big bang , neutron stars , white dwarf stars and various exotic stable configurations of quantum fields , such as non - topological solitons .
gravitational microlensing is one of the ways to detect these objects .
however , this method only gives us the information about the mass , distance and velocity of them and nothing more .
+ heavier , relatively faint objects such as old white dwarfs and mini black holes , are the most important candidates for this bdm .
if there are primordial mini black holes produced in the big bang before bbn(big bang nucleosynthesis ) , they would not contribute to @xmath20 .
+ @xmath21 is required by a star to start to shine as a star by igniting thermonuclear fusion .
brown dwarfs are not completely dark because they radiate faint thermal radiation .
however , the typical mass of these machos should be @xmath22 , which is much larger than the brown dwarf mass .
in addition , white dwarf has a mass which is suitable for dark matter but there is not enough number of observed white dwarfs .
hence , bdm is not as good candidates for dark matter as nonbaryonic dark matter .
hot dark matter ( hdm ) and cold dark matter ( cdm ) is considered as a nonbaryonic dark matter .
other more special options have been considered , such as for example warm dark matter ( wdm ) . +
dark matter particles that are called hdm , are relativistic at the time of matter - radiation equality time .
this requirement implies that their mass has to be lower than about 100 ev .
their thermal velocities , when structure formation begins are large because of the small mass . in contrast , cdm has large mass so its velocities are negligible .
candidates between hot and cold are called warm dark matter .
+ effect of the pull of gravity causes to form galaxies and galaxy clusters from early homogeneously distributed . this is called structure formation . today
the best way to differentiate between hdm , wdm and cdm is through the observed large - scale structure , i.e. cmb which shows the seeds of structure .
we show in fig .
( 1.3 ) the results of a simulation of the halo of dark matter around the milky way and two other galaxies . for cdm ,
there is more substructure and satellites around the galaxy , while their formation is suppressed for wdm .
according to observations , there is an order of magnitude less satellites observed around the milky way than predicted by cdm models .
however , the observations are not complete , and the discrepancy may also be due to other causes than wdm . ]
the most common candidates for non - baryonic dark matter are weakly interacting massive particles , or wimps @xcite .
cold dark matter has a most important candidate which is wimps . because they have neutral charge and they are long lived particles . only weak force and gravitational force effect them .
large amount of wimps are produced and then the interaction rate of wimps was decreasing so decoupling increasing between them in the thermal bath of the early universe when cools down , like neutrinos , but are much heavier , so that they are a form of cdm .
after the decoupling , their relic density remains constant as consistent with experiments .
this is called wimps miracle .
furthermore the interactions of some dark matter candidates are stronger or weaker .
for example , gravitinos have only gravitational interactions , while timps ( technicolour interacting massive particles ) have much stronger interactions .
neutralino ( bino+wino+2higgsino ) lightest supersymmetric particle is the best candidate for wimp in minimal supersymmetric standard model ( mssm)@xcite .
the other candidate for wimp is the lightest kaluza klein ( lkp ) which is coming from universal extra - dimensions ( ued ) model
. for the hdm neutrinos are candidates .
energy density today is dominated by their rest masses so to make a significant contribution to the density parameter , neutrinos should have a rest mass about 1 ev . the present upper limit to hdm in the form of massive neutrinos leads to the conservative mass limit from wmap data .
@xmath23 therefore the maximum contribution of neutrinos is @xmath24 , at most the same order of magnitude as baryonic matter .
if neutrinos were the dominant form of matter , there would be a lower limit on their mass from constraints on the phase space density , called the tremaine - gunn limit .
essentially , in order to achieve a certain rotation velocity for galaxies , you need a certain amount of mass inside a given volume , and the pauli exclusion principle constrains the number of particles you can pack inside a given volume .
even though we know that neutrinos are a subdominant component of dark matter , the tremaine - gunn limit applies to any fermionic dark matter candidate , even if its distribution is not thermal .
( there is no similar lower limit on the mass of a bosonic dark matter particle . )
most of the matter in the universe has to be cdm because of data on large scale structure combined with structure formation theory .
wimps ( weakly interacting massive particles ) are the main candidates of cdm .
dark matter candidates are with masses in the 100 gev to tev range .
particles with gev to tev masses and weak - scale interaction cross sections are known as weakly interacting massive particles ( wimps ) .
there are also another candidates for dm . however , wimp dark matter is compelling for two main reasons : 1 . in the early universe ,
when wimps were in thermal equilibrium annihilated with each other with a weak scale cross section . as the universe expanded and has frozen out , wimps cooled to obtain a relic density today which is equal to the measured dark matter density . + as mentioned before that this is known as the wimp miracle .
the other reason is that wimp dark matter are present in many of the favored extensions of the standard model.for example , the minimal supersymmetric standard model ( mssm ) has a favourite wimp candidate .
by changing the law of gravity , it could be possible that the observations are explained .
until the dark matter particle is detected , there is some room for uncertainty .
the problem for modified gravity is that there are so many different observations for dark matter , in different physical systems : motions of stars in galaxies , motions of galaxies in clusters , gravitational lensing , large - scale structure , cmb anisotropies and so on .
gravity has to be adjusted in a different manner for these different observations , and the resulting models are somewhat contrived
. expressed another way , the dark matter scenario is very predictive : the simple hypothesis of a massive particle with weak couplings to itself and to the standard model particles explains a number of disparate observations and has made successful predictions . + in 1919 and 1922 respectively by weyl and eddington @xcite gravitational action firstly studied .
however , until 1960s , there was not any indications of complicating the gravitational action .
because of cosmological problems such as the dark energy problem , dark matter problem , this type of modification is needed to fix the problems . at @xmath0 theory ,
the main idea is to take action as a function instead of a constant curvature such as : @xmath25\ ] ] by using the @xmath0 function , it provides usage of many different properties , which are very important for explaining unknown matters such as dark matter and dark energy .
for instance , considering f as a series expansion i.e
@xcite @xmath26 where @xmath27 and @xmath28 coefficients have the appropriate dimensions .
so it can be seen that a number of interesting terms appear .
@xcite without matter action is given above , now after adding matter on the action , our @xmath0 action is @xmath29.\ ] ] varying the action with respect to @xmath30 provides @xmath31 where @xmath32 , @xmath33 and @xmath34 $ ] so @xmath35 @xmath36 @xmath37 @xmath38 in which a prime means d / dr and its variation with respect to @xmath30 yields , after some manipulations and modulo surface terms , the field equation @xmath39 where @xmath40 is the covariant derivative associated with levi - civita connection of the metric , and @xmath41 .
these equations above are fourth order partial differential equations in the metric . the alternative name `` fourth order gravity '' used for this theories because in the first two terms on the right hand side fourth order derivatives appear .
furthermore , the contraction of the above equation yield the trace equation @xmath42 where @xmath43 the trace equation can be used to simplify the field equations and then keep it as a constraint equation .
therefore @xmath44 the trace of this equation is @xmath45 where @xmath46 here we can see that when @xmath47 , not anymore says that @xmath48 or constant .
another important thing is that when @xmath49 and @xmath50 , it reduces to @xmath51 maximally symmetric solutions is affected by f function , and they will be minkowski , de sitter or anti - de sitter .
hence , we can now write the field equations in the form of einstein equations as @xmath52}{2f_{r}}+\frac{[\nabla_{\mu}\nabla^{\nu}f_{r}-g_{\mu\nu}\box f_{r}]}{f_{r}}\ ] ] after some algebra the above equation , takes the form @xmath53 where @xmath54 . for solving above equation
we need a metric and its components@xcite . here
we are using spherically symmetric metric with radial components b(r ) and x(r ) : @xmath55 ricci scalar of this metric can be written as @xmath56 for empty space if we solve the einstein equation with the form of @xmath0 gravity , two independent field equations are found : @xmath57 @xmath58 where the prime means d / dr .
after that are choose our ansatz with constants @xmath59 , @xmath60 and @xmath61 as @xmath62 constans are bounded by two constraints as : @xmath63b_{0}.\ ] ] after that @xmath64 is obtained . where @xmath65 this will be the well known solution of schwarzschild when @xmath66 or n goes to zero . after using those constraints and definitions , an asymptotic form of @xmath0 where @xmath67= constant of integration is found as @xmath68 in the paper of a model of @xmath0 gravity as an alternative for dark matter in spiral galaxies they took @xmath69
when n is going to zero and action goes to @xmath70 . for the application of this in the rotation curve ,
geodesic equation is used in the weak field approximation .
the test particle is rotating around the central mass and obtains as @xmath71 where the dot means d / dt . after using the affine connection which we found by using metric : @xmath72 @xmath73 velocity for a test particle is obtained as @xmath74 after considering some assumption for the large distances from center of galaxy , such as using @xmath75 where @xmath66 is smaller than 1 , one finds that @xmath76 for taking @xmath77 where @xmath78 is the proportional coefficient with mass inverse dimension and m is mass of galaxy , hence asymptotic velocity is found @xmath79 where @xmath66 and @xmath78 depend on experimental datas .
for instance , if this values taken as @xmath80 , the velocity is found exactly @xmath81 and this value is consistent with tully - fishers relation .
@xcite furthermore , another good testing is made by using the modified newtonian dynamics ( mond ) .
@xmath82 for the velocity of typical galaxy .
there are many difficulties of this theory .
one example which gained a lot of publicity a few years ago is the bullet cluster .
a solution of the dark matter conundrum in terms of a new elemantary particle is quite successful to explain the bullet cluster , while explanations in terms of modified gravity models encounter serious difficulties .
another problem in this way of solution is that ansatz is choosen by hand and not naturally comes out .
wimps and many other dark matter candidates are in thermal equilibrium in the early universe and decouple once their interactions become too weak to keep them in equilibrium .
those particles are called thermal relics , as their density today is determined by the thermal equilibrium of the early universe .
when the annihilation processes of wimps into sm particles and vice versa happend at equal rates , there is a equilibrium abundance .
when the universe cooled down and the rate of expansion of the universe h exceeds the annihilation rate , the wimps effectively decouple from the remaining sm particles .
so the equilibrium abundance drops exponentially until the freeze - out point.the abundance of cosmological relics remained almost constant until today @xcite .
if the candidate is stable or has a long lifetime , the number of particles is conserved after it decouples , so the number density falls like @xmath83 .
specifically , we use the lee - weinberg equation @xcite .
it describes annihilation and creation of @xmath84 particles @xcite .
@xmath85 with @xmath86 , the number density of the relic particles in equilibrium .
furthermore hubble constant make expanding universe effect .
the thermal average of the annihilation cross section @xmath87 multiplied with the relative velocity v of the two annihilating @xmath84 particles .
the first ( second ) term on the right - hand side ( rhs ) of the equation describes the decrease ( increase ) of the number density due to annihilation into ( production from ) lighter particles .
kinetic equilibrium is satisfied by the lee - weinberg equation for @xmath84 now we use hubble constant effect which provide universe expansion by considering the evolution of the number of particles in a portion of comoving volume given by @xmath88 we can then introduce the convenient quantity @xmath89 such that @xmath90 .
in addition , since the interaction term usually depend explicitly upon temperature rather than time , we introduce the @xmath91 , the scaled inverse temperature . during the radiation dominated period of the universe , thermal production of wimps
takes also place . for this time
, one can find the expansion rate @xmath92 and then @xmath93 where @xmath94 and where @xmath95 refer the number of intrinsic degrees of freedom such a like that spin or color for @xmath84 particle .
hence we can write the last format of our equation a @xmath96 where @xmath97m^{3}}{4 \pi^{2 } x^{3}}\ ] ] after all little tricks , our lee - weinberg equation can be recast as @xmath98 where we defined @xmath99}{4\pi^{2}}\sqrt{\frac{45}{4\pi^{3}}}\frac{g_{\chi}}{\sqrt{g_{\chi}}}mm_{pl}<\sigma v>=0.055\frac{g_{\chi}}{\sqrt{g_{\chi}}}mm_{pl}\sigma_{0}\ ] ] where @xmath100 is eulerriemann zeta function to integrate the lee - weinberg equation , we need to have an expression for the equilibrium number density in comoving volume.once the particle is non - relativistic , the difference in statics is not important .
the general equation of equilibrium number density in comoving volume is @xmath101}\int_{0}^{\infty}\frac{y^{2}}{1+\exp^{\sqrt{x^{2}+y^{2}}}}dy\ ] ] for the nonrelativistic case at low temperatures @xmath102 one obtains , @xmath103}\frac{\pi}{2}x^{3/2}\exp^{-x}\ ] ] at great value of temperatures , dark matter quickly is destroyed by its own antiparticle . shortly after that , until @xmath104
, t has dropped too much which is under the value of @xmath105 , also the number density of dark matter drops exponentially .
the temperature at which the particle decouples ( the time when the number of particles reaches this constant value ) is called freeze - out temperature @xmath106 .
hence the number density per comoving volume becomes almost constant because dark matter particles are no longer able to annihilate efficiently .
an approximate solution for the relic abundance is given by @xmath107 .
where @xmath108 and @xmath109 is the so called freeze - out temperature and approximately , for the typical case @xmath110,one has that @xmath111 is a solution of @xmath112 we can then finally calculate the contribution of @xmath84 to the energy density parameter finding a well know result @xmath113 it is intriguing that so called `` wimps''(e.g .
the lightest supersymmetric particle ) dark matter particles seem to reproduce naturally the right abundance since they have a weak cross section @xmath114 and masses @xmath115 .
this observations is called the `` wimp miracle '' and typically considered as an encouraging point supporting wimps as dark matter candidates . and @xmath116 as function of x for m = 100 gev , @xmath117 , g = 90 ( constant ) , @xmath118 , @xmath119 ] ] the lw eqn .
can be solved by approximately analytic calculation or numerical plotting by using mathematica.we have plotted those equation by using both of the ways . as you can see fig .
the termal production , is the most traditional mechanism but that many other mechanisms have been considered in the literature such as non thermal production of very massive particles(so called wimpzillas ) at preheating or even right - handed neutrino oscillations .
the observational evidence for dark matter is overwhelming and now we are entering one of the perhaps most exciting periods in physics history where a wide array of experiments may have the potential to unveil its nature.in this thesis i review some important concepts about dark matter .
one of the best motivated candidates for dark matter so far is the wimps which i mentioned before . in this thesis
the focus was on dwarf spheroidal galaxies , promising targets that are assumed to be strongly dark matter dominated systems due to their high mass - tolight ratios up to few hundred solar units .
dwarf galaxies of the milky way have the advantage of relative proximity and low astrophysical background as they do not lie within the galactic plane .
great importance was given to studying the properties of these systems and how their dark matter profiles can be derived from kinematical studies .
however a precise estimation of the density distribution is a challenging task and due to too little data available in particular the shape of the dark matter halo inner regions remains poorly known .
cusped profiles suggested from simulations like the navarro - frenk - white profile are not ruled out in general although observations seem to favor profiles with a flat inner core .
the two investigated dwarf spheroidal galaxies sagittarius and canis major are particularly attractive targets as they are the closest dwarf galaxies discovered so far .
the uncertainty concerning their inner region is reflected in the plots of sensitivity limits for both investigated dwarf galaxies , where the distances between the two lines of a cored and a cusped profile are about one order of magnitude .
better opportunities for a significant dark matter detection could be obtained for instance by the next generation ground - based cta ( cherenkov telescope array ) experiment with bigger effective areas due to many cherenkov telescopes in the order of 50 and also icecube experiment . here
, stereoscopy will be possible for small energies .
moreover the space based ray telescope fermi will yield interesting results in the future.apart from sagittarius and canis major , there could be promising faint dwarf galaxies that are not discovered yet
. however , there should be as many undiscovered dwarf galaxies in the other direction , too , which , when discovered , will provide interesting targets for the search of dark matter .
in chapter 2 we gave a simple example of @xmath0 extended gravity in which the effect of dm is not due to exotic sources but due to the wrong choice of lagrangian .
einsteins model , which extends that of newton is characterized by the specific choice @xmath120 . in the @xmath0 model
this has been modified to different polynomial powers of r. as an example the choice @xmath121 where @xmath122 , describes asymptotic behaviours of flat rotation curves and experimental results of tully - fisher ( i.e. @xmath123 mass ) .
for n=0 the model reduces to the einsteins general relativity model which fails to explain flat rotation curves unless supplemented by exotic sources .
kapteyn , ap .
j. * 55 * , 302 ( 1992 ) .
f. zwicky , helv .
acta * 6 * , 110 ( 1933 ) .
a. slyz and t. kranz , mon . not .
soc . * 346 * , 1162 ( 2003 ) .
j. bahcall and x. fan , astrophys .
j. * 504 * , 1 ( 1988 ) .
r. g. carlberg , phys .
rept . * 307 * , 67 ( 1998 ) .
d. j. eisenstein , astrophys .
j. * 633 * , 560 ( 2005 ) .
hinshaw , astrophys .
* 148 * , 135 ( 2003 ) .
nasa s wilkinson microwave anisotropy probe ( wmap ) , http://map.gsfc.nasa.gov/. particle data group , j. phys .
g. * 33 * , 224 ( 2004 ) .
g. bertone , d. hooper and j. silk , phys . rep .
* 405 * , 279 ( 2005 ) .
h. e. haber and g. l. kane , phys
. rep . * 117 * , 75 ( 1985 ) .
wmap collab . and
d. n. spergel , astrophys . j. suppl .
* 170 * , 377 , ( 2007 ) .
e. w. kolb , d. j. h. chung and a. riotto , arxiv : hep - ph/9810361 .
k. g. begeman , a. h. broeils and r. h. sanders , mnras * 249 * , 523 ( 1991 ) .
f. zwicky , ap .
j. * 86 * , 217 ( 1937 ) .
r. h. cyburt and h. richard , phys .
d * 70 * , 023505 ( 2004 ) . b. w. lee and s. weinberg , phys .
* 39 * , 165 ( 1977 ) .
a. s. eddington , cambridge university press , cambridge ( 1923 ) .
h. weyl , ann . phys . * 59 * , 101 ( 1919 ) . s.asgari and r.saffari , appl .
. res . * 2 * , 99 ( 2010 ) .
t. p. sotiriou and v. faraoni , rev .
phys . * 82 * , 451 - 497 ( 2010 ) .
s. capozziello and m. de laurentis , annalen der physik * 524 * , 545 ( 2012 ) .
s. nojiri and s. d. odintsov , arxiv : 0807.0685v1 . s. nojiri and s. d. odintsov , phys . rev .
d * 68 * , 123512 ( 2003 ) .
s. nojiri and s. d. odintsov , gen .
. grav . * 36 * , 1765 ( 2004 ) .
r. saffari and s. rahvar , phys .
d * 77 * , 104028 ( 2008 ) .
p. gondolo and g. gelmini , nucl .
b * 360 * , 145 ( 1991 ) .
e. w. kolb and m. s. turner , addison - wesley ( 1990 ) .
p. di bari , some notes on neutrino physics and particle cosmology , http://152.78.192.50/ pdb1d08/neupc1011.pdf ( 1990 ) . | the identity of dark matter is one of the key outstanding problems in both particle and astrophysics . in this thesis ,
i review some candidates of dark matter , especially wimps ( weakly interacting massive particles ) which is one of the best candidate so it is called that wimps miracle .
in addition of this , there are also some theories of modification of gravity , by changing the law of gravity , it could be possible that the dark matter observations are explained . until the dark matter particle is detected , there is some room for uncertainty .
so we should consider every part of the problem and solve it .
dark matter problem is covering a large area so every possibility is important .
so @xmath0 gravity is also reviewed in this thesis and some theories are considered as a possible solution of dark matter problem .
finally we highlight that , even in the case of wimps or another particles solution , @xmath0 gravity is also can be used for this problem .
however , last words will be said by experiments .
* keywords * : dark matter , weakly interacting particles(wimps ) , @xmath0 gravity , cosmology .
karanlik maddenin gizemi hem astrofizik hem de paracik fizii aisindan ok nemli bir problemdir .
bu tezde , karanlik madde adaylarini , zellikle wimps mucizesi olarak bilinen ve en gl aday olan wimps(zayif etkileen ktleli paracik)i inceledim .
karanlik maddenin gizemini zmeye alian bundan baka teoriler de vardir , rnein gravitasyonun modifiye edilmesi gibi .
karanlik madde paraciklari deneylerde bulunana kadar , bu gizem srecektir . biz
bu gizemi her aidan inceleyip zm retmemiz gerekir nk ok geni bir alani kapsayan bu alan iin her bir ihtimal bile ok nemlidir .
karanlik maddeyi aiklamaya aday olan @xmath0 gravitasyon teorisini de bu tezde inceledik ve zm retmeye alitik .
sonu olarak , gerek wimps olsun gerekse baka paraciklar olsun , @xmath0 gravistasyon teorisi de gzel bir alternatiftir ve kullanilabilir .
tabii hangisinin doru olduunu deneyler onaylaycak .
+ + + + + * anahtar kelimeler * : karanlik madde , zayif etkileen ktleli paraciklar(wimps ) , @xmath0 gravistasyon teorisi , kozmoloji . to my family
i would like to express my deep gratitude to prof.dr.mustafa halilsoy , my supervisor , for his patient guidance , enthusiastic encouragement and useful critiques of this research work .
i would also like to thank assoc.prof.dr.izzet sakalli , for his advice and assistance in keeping my progress on schedule .
my grateful thanks are also extended to prof.dr.zay grtu , asst.prof.dr.habib mazharimousavi and asst.prof.mustafa riza .
i would also like to thank pasquale di bari , who was my supervisor at university of southampton for his advice on the dark matter production .
i would like to thank my friends in the department of physics , the gravity and general relativity group : kiymet emral , morteza kerachian , tayebeh tahamtan , marzieh parsa , yashar alizadeh for their support and for all the fun we have had during this great time .
i wish to thank also other my friends for their support and encouragement throughout my study .
finally , i would like to thank my family . |
the most important chemical transitions for quasar absorption line systems at low redshift ( e.g. , @xmath0@xmath11 ) still lay in the ultra - violet spectral range .
a limited number of the most uv - bright quasars have been observed with the hubble space telescope ( _ hst_)/ space telescope imaging spectrograph ( stis ) , which has provided a detailed view of absorption systems along the lines of sight .
these data are available from the multimission archive at space telescope ( mast ) . however , in order to compile statistics of absorbers or to study particular systems , it is necessary to identify the spectral features .
this can be complicated , particularly in the forest region , and when the available spectral coverage is limited .
in the course of our studies of the forest and metal - line systems we have studied in detail eight of the highest quality _
spectra , obtained with the e230 m grating . in this paper we present these spectra with a list of @xmath2 absorption features and our suggested line identifications .
in particular , metal - line system identifications are necessary in order to remove contaminants from the forest . in a recent paper @xcite , we measured statistics of absorption in the forest at @xmath3 using 74 low resolution _
quasar spectra . as a check , and particularly to assess our ability to separate metal lines from the forest , we also used the available higher resolution e230 m _
hst_/stis spectra . in the present paper ,
we present the relevant data and line identifications which were used in that paper .
we expect that this catalog will facilitate future studies of metal - line absorption systems as well as of the forest .
our catalog includes damped absorbers ( dlas ) , lyman limit systems , and forest clouds .
the dla and lyman limit systems are expected to have strong absorption @xcite , while the forest clouds may have weak and other low ionization transitions detected ( such as and ) @xcite .
some forest clouds have detected and/or , without associated low ionization absorption @xcite .
the catalog that we will present will allow for future studies of the various types of absorbers . in [ sec :
data ] we describe _
quasar spectra and our reduction and continuum fitting procedures .
we also outline our line detection and measurement algorithms and line identification strategies .
the normalized echelle spectra are presented in [ sec : results ] , along with a brief description of metal - line systems detected toward each quasar .
system plots , showing all detected transitions for each system , are also presented there . in [ sec : summary ] we present a general summary of the systems presented in this catalog .
we selected the eight @xmath4 quasars observed with the _
e230 m grating which had a signal to noise ratio , @xmath5 , over a reasonable fraction of the spectrum .
hst_/stis e230 m spectra have a resolution @xmath6 , with two pixels per resolution element .
our sample is biased towards quasars that have lyman limit systems , since most of the stis observations were conducted in order to study those particular known systems in detail .
several settings are possible covering different wavelength ranges . for seven of the eight quasars in our sample , @xmath7 - @xmath8 is covered .
the quasars and the relevant observational details are listed in table [ tab : tab1 ] . specifically , this table lists quasar redshifts , wavelength coverage , @xmath9 per pixel at 2350 and 2750 , primary investigator for the original observation , and proposal i d .
the data were reduced using the stis pipeline @xcite .
they were combined by simple weighting by exposure time ( as in @xcite ) , rather than by , e.g. , inverse variance methods that are often employed .
a bias could be introduced by the latter method , since pixels with smaller counts would be weighted more heavily .
this bias is significant only in cases such as these stis spectra , where individual exposures have very small @xmath9 , and is not typically important for reduction of ground - based high resolution echelle spectra .
wavelengths were corrected to the heliocentric reference frame . when the same quasar was observed multiple times
, there was often a small shift in wavelength between the spectra , due to the intentional shifting of the echelle angle of the instrument . in this case , we chose not to smooth the data by interpolating , but instead chose wavelength bins from one exposure and combined the flux from other exposures into the nearest bin .
this results in a slightly decreased effective resolution .
continuum fits used the standard iraf sfit task .
features were objectively identified by searching for an unresolved line at each pixel ( as employed in @xcite ) , and applying a @xmath10 criterion for detection .
however , we found by inspection that some of these formal detections are not likely to be real . in some cases , only 1 pixel had significant absorption , and in other cases it appeared that correlated noise biased the measurement .
features that are broad and shallow can be @xmath11 5@xmath12 and yet look unconvincing because they are sensitive to the continuum level .
such spurious features , estimated to be 10% of the total number of detections , were eliminated from consideration .
we define a feature as all the pixels around a 5@xmath12 detection that have not recover to the continuum after smoothing the spectra using an equation on p.56 of @xcite .
this definition means that many features are clearly blends of well separated lines , and hence a feature can have more than one identification , each of which may be secure because they refer to different blended lines .
wavelength that we give for a feature is the flux - weighted central value , the mean of the wavelengths where the smoothed flux drops below the continuum , with weighting for the fraction of the photons that are absorbed at each pixel .
we attempted to identify all features in each of the quasar spectrum , and the transitions that were identified one or more times are listed with their vacuum wavelengths and oscillator strengths in table [ tab : tab2 ] . for our line identification we used the following procedure : 1 .
examine the spectrum at the expected positions of possible galactic lines , and mark all possible detections .
search for the strong resonant doublet transitions : mg0.1emii @xmath13 , si0.1emiv @xmath14 , c0.1emiv @xmath15 , , and o0.1emvi @xmath16 .
in addition to these , we also search for low ionization gas through the combination @xmath17 and @xmath18 , which can be used in place of mg0.1emii @xmath13 when the latter is not covered @xcite .
check positions of other transitions that could be detected for systems found through the doublet search , including the lyman series lines . if a system is ambiguous with the doublet alone , and other lyman series lines were used to assess its reality .
4 . having completed these steps , all features that remain without plausible identifications are considered to be lyman series lines .
begin by assuming that the highest order lyman series line is the correct identification and search for all of the stronger ( lower order ) lyman series lines corresponding to the same redshift .
if reasonable ratios are found , mark these as possible identifications . otherwise continue assessing the possible identifications of the feature in question until reaching .
if no identification , including a identification , is plausible the feature is left unidentified .
7 . if more than one identification is reasonable ( or if there is likely to be a blend for which more than one system contributes significantly ) , this is noted .
figure 1a is a sample of a portion of the normalized spectrum of pg@xmath19 , shown in the printed version .
we present all other normalized quasar spectra in figures 1b8d ( see the complete version ) , which are available electronically at http://www.blackwellpublishing.com / products/. similarly , our listings of line identifications are presented electronically in table [ tab : tab3 ] , with some sample listings appearing in the printed version . in this table , quasars are ordered by right ascension and @xmath10 features by increasing wavelength within a given quasar entry . in table
[ tab : tab3 ] we list , for each @xmath10 feature that we did not reject as unconvincing , the observed vacuum wavelength ( @xmath20 ) , the observed frame equivalent width and error ( @xmath21 and @xmath12(@xmath21 ) ) , and the significance level ( @xmath22 ; observed equivalent width/ error in equivalent width ) .
it also gives our favored line identification for each feature ( transition and redshift ) along with notes that indicate alternative and/or additional identifications . if the identification can not be confirmed with another line , or is uncertain for some other reason , a `` ? ''
is listed after the preferred transition i d .
we do not give column densities since there are best done on a system by system basis using velocity plots to ensure that the columns for different ions are representative of the same gas .
the following subsections present the eight quasars along with discussion of noteworthy metal - line systems found in their spectra .
figure 9 presents an example `` system plot '' for the @xmath23 system toward pg@xmath19 .
the transitions detected at @xmath10 for all of the metal - line systems are plotted in velocity space in the electronic versions of figures 964 ( see the complete version ) .
the quasar emission redshifts , taken from the nasa / ipac extragalactic database , are listed in the heading of each subsection .
this quasar was studied as part of the quasar absorption line key project , using g270h grating of _ hst_/faint object spectrograph ( fos ) , with resolution of @xmath25=1300 .
@xcite found possible metal - line systems at @xmath26 , 0.9400 , 0.9676 , 1.0724 , 1.3389 , 1.3426 and 1.3868 .
they also found tentative suggestions of metal - line systems at @xmath27 and 1.4952 .
is detected in a keck / hires spectrum at @xmath23 , 0.7290 , 1.0480 , 1.3250 and 1.3430 @xcite , and these five systems were modeled ( including constraints from this stis coverage ) by @xcite .
@xcite observed this quasar with _
hst_/stis to facilitate a statistical study of the forest .
this spectrum covers the forest over almost all of its wavelength coverage , blueward of @xmath28 .
the line is covered blueward of @xmath29 , blueward of @xmath30 , and blueward of @xmath31 .
there is also limited redshift coverage of higher order lyman series lines . along this line of sight are five absorbers .
one of them is a dla , two are strong absorbers , and the other two are multiple - cloud , weak absorbers .
three certain , and two possible systems are also detected .
the metal - line systems toward this quasar are plotted in figures 918 .
the @xmath23 system is a strong absorber .
based upon only a low resolution _ hst_/fos spectrum , obtained in spectropolarimetry mode @xcite ,
it is not clear whether this is a dla or whether it has multiple undamped components @xcite . in the _ hst_/stis spectrum ,
c0.1emiv @xmath15 are detected , but both transitions suffer from blending .
the @xmath32 is blended in its blue component with from a system at @xmath33 .
an unidentified blend appears on the red wing of @xmath34 and its blueward component also has a small contribution from @xmath35 at @xmath36 .
@xmath37 is detected , but the blueward member has a blend to its blue , probably with .
strong @xmath38 is detected , as is @xmath39 .
the strong feature redward of @xmath38 is probably at @xmath40 .
in addition to and , and were detected in the keck / hires spectrum @xcite . the unusual multiple - cloud , weak absorber at @xmath41 is detected only in @xmath38 and @xmath42 in the _ hst_/stis spectrum .
c0.1emiv @xmath15 is covered at a high sensitivity , yet it is not detected to a @xmath43 rest - frame equivalent width limit of @xmath44 .
this makes it `` deficient '' , possibly indicative of a low level of star formation so that a corona would be weak or absent in its very red , barred spiral galaxy host @xcite .
the @xmath36 system was found not to have metal lines detected in the low resolution _
hst_/fos spectrum .
the detections of , , , si0.1emiv @xmath14 in the _
hst_/stis spectrum are consistent with those limits .
unfortunately , the stis spectrum does not cover c0.1emiv @xmath15 , which would provide useful constraints on the system s physical conditions .
based upon photoionization modeling @xcite it would appear that there is a significant blend with @xmath45 , which is most likely .
the @xmath46 system is just above the borderline to qualify as a strong absorber .
it has six weak components in mg0.1emii @xmath13 spread over @xmath47 @xcite .
there is a partial lyman limit system detected in an _
hst_/fos spectrum @xcite . in the _
hst_/stis spectrum and are detected from the strongest four of these six components .
@xmath481207 is also detected over the full velocity range , but it suffers from a blend with galactic @xmath49 to the blue .
there may also be blends redward of this , based upon the photoionization models @xcite , but we were unable to provide identifications besides possible .
unfortunately , c0.1emiv
@xmath15 , which was detected in the _
hst_/fos spectrum @xcite , is not covered in the stis spectrum .
there is a possible weak detection of @xmath481239 but it can not be confirmed using @xmath481243 because it is in a noisy part of the spectrum .
a strong o0.1emvi @xmath16 is detected , but @xmath50 is blended to the red with from the system at @xmath51 , and to the blue with possible .
also , @xmath52 must have a blend , at least on the red wing of its strong component , since that profile is too strong relative to @xmath50 at that velocity .
this blend could be @xmath50 from the @xmath51 system .
if so , these systems would be `` line - locked '' .
we find the evidence for metal lines in the @xmath51 system to be inconclusive .
is not detected in the redward portion range of the possible @xmath53 feature , so not all of that absorption can be at this redshift .
two components of @xmath54 are possible , but they do not align perfectly with the minima of the profile and can not be confirmed in any other way . also , o0.1emvi
@xmath16 may be detected .
the alignment is reasonably good , though the feature redward of the @xmath50 component is likely to be @xmath52 from the system at @xmath46 .
the @xmath55 system is also just at the border between strong and multiple - cloud weak absorption . like the @xmath46 system
, it has a partial lyman limit break in the _
hst_/fos spectrum @xcite . and
are detected in the _
hst_/stis spectrum along with , , , .
there is a feature at the expected position of @xmath56 , which appears to be too strong relative to the other transitions in this system @xcite .
we believe this feature could be at @xmath57 .
is not covered in the stis spectrum , but strong absorption is detected in the fos spectrum . appears to be detected , but the @xmath58 transition is blended , probably with at @xmath59 .
there is an system , with detected , , and at @xmath60 .
another weak , broad doublet at @xmath61 is also detected in and , as well as in .
the feature in the @xmath54 panel , at @xmath62 , is at least partially @xmath18 at @xmath55 ( as is the stronger feature to its red ) . at @xmath63
the lyman series is detected down to , which is the last member covered in the _
hst_/stis spectrum .
there is a possible doublet at this redshift , but the feature at the position of @xmath50 transition does not match the profile of that at the position of @xmath52 . the system at @xmath33 is another example of associated absorption that appears slightly redward of the quasar emission redshift . it is detected in and .
the shows the same components as , but clearly has a blend to the red , probably with . is heavily blended with @xmath32 from the @xmath23 system .
@xmath53 is also detected in three components , though the blueward one is affected by a data defect .
this quasar was observed with _
hst_/stis by @xcite in order to study a sub - dla at @xmath65 , for which they also have optical coverage .
their study of this system , particularly focused on molecular hydrogen , is published as @xcite .
in addition , @xcite and @xcite report on six systems along the line of sight ( at @xmath66 , 1.416 , 1.602 , 1.674 and 1.697 ) , based upon the _ hst_/stis spectrum in conjunction with vlt / uves data . is covered over the entire _
hst_/stis spectrum , up to @xmath67 , and is covered up to @xmath68 . there is more limited coverage of all other lyman series transitions down to the lyman limit of the quasar . the _ hst_/stis spectrum is dominated by the sub - dla at @xmath65 and its many detected metal and molecular hydrogen transitions . in addition , there is one system , four definite systems , and three possible systems .
a couple of these systems are associated .
system plots for this quasar are given in figures 1927 .
a two component doublet is detected at @xmath69 , which also has and detected .
is also probably detected in the redward component , but this ca nt be confirmed because @xmath70 is blended with an unknown line , probably .
the sub - dla at @xmath71 is a very complex system , spanning @xmath72 for absorption in many low ionization transitions .
it has many molecular hydrogen lines detected .
our identifications for the molecules follow exactly those of @xcite .
the saturated region of the sub - dla profile spans @xmath73 .
a separate component at @xmath74 is likely also to be , but with no associated metals in the _
hst_/stis spectrum .
many neutral species of and are detected in a single , narrow component at @xmath75 , presumably associated with the bulk of the .
features at other velocities seen in the neutral transitions panels in figure 20 are not related to this system ( see line identifications in table [ tab : tab3 ] ) .
@xmath76 is blended with at @xmath77 , and can not be measured , and no other strong transitions are covered .
many singly ionized transitions are detected for the system , both in the @xmath78 component and at many other velocities to the blue .
these match the and from the vlt / uves spectrum @xcite .
strong absorption is detected in many components , spanning the full velocity range .
si0.1emiv @xmath14 is detected as well , but there appears to be significant blending with @xmath79 , because @xmath80 is much weaker at some velocities .
c0.1emiv @xmath15 is also detected in the vlt / uves spectrum . is covered , but not detected , and o0.1emvi @xmath16 is not covered .
an doublet detected in the _
hst_/stis spectrum at @xmath81 has its @xmath52 member blended to the blue with at @xmath82 .
c0.1emiv @xmath15 is detected at the same redshift in the vlt / uves spectrum @xcite .
the corresponding , relatively weak ( unsaturated ) line lies blueward of a larger saturated feature . may also be detected .
there is a possible system at @xmath83 , as reported by @xcite , however neither member of the doublet aligns with the and , which are also detected in the stis spectrum .
there could be associated with this system , but it is badly blended with a probable line , thus its alignment can not be verified . therefore , there is not convincing evidence for detected metals in this system .
the system at @xmath84 is well aligned and is detected in and in the _
hst_/stis spectrum . , and
possibly , is detected in the vlt / uves spectrum @xcite .
there is a possible system at @xmath85 , though its identity is somewhat uncertain .
@xmath52 is blended with a possible line , such that the profiles of the doublet members can not be compared .
is not detected at the position , however is detected , with about the same profile shape and equivalent width as the .
thus , this could be a fairly typical associated absorber .
there is an system at @xmath86 detected in the _
hst_/stis spectrum .
@xmath52 is blended to the red with a possible line .
is detected , as are higher order lyman series lines , but the latter are blended with other transitions ( see table [ tab : tab3 ] ) .
c0.1emiv @xmath15 and are detected in the vlt / uves spectrum @xcite .
the associated absorber at @xmath87 also has @xmath53 and probably @xmath88 detected .
@xcite also detected , si0.1emiv
@xmath14 , c0.1emiv @xmath15 , and in their vlt / uves spectrum .
the profile is not black , again a signature of associated absorbers .
a possible system at @xmath89 can not be confirmed by other features in the _
hst_/stis spectrum .
although is covered , it is blended with galactic @xmath49 . if real , this system would be significantly redshifted relative to the emission redshift of the quasar .
this quasar was observed by _
hst_/fos as part of the quasar absorption line key project , but they had difficulty in identifying many of the lines due to uncertain continuum fitting @xcite .
they found evidence for an extensive metal - line system separated into subsystems at @xmath91 , 0.9277 and 0.9342 , even at low resolution .
@xcite modeled this system based on the _
hst_/fos data , along with keck / hires data @xcite .
@xcite refined these models once the _
hst_/stis spectra were available .
@xcite observed this quasar with _
hst_/stis to facilitate a detailed study of the systems at @xmath92 .
hst_/stis spectrum covers the forest blueward of @xmath93 , but does not cover any higher order lyman series lines .
there are only three metal - line systems found in the _
hst_/stis spectrum : the extensive metal - line system at @xmath92 mentioned above , a system , and an associated system .
figure 2832 presents system plots for these metal - line systems , detected toward pg@xmath94 .
there is a weak doublet at @xmath95 .
the @xmath32 is blended with @xmath79 from the subsystem at @xmath91 , and also to the red with an unknown blend , probably .
the large feature blueward of the @xmath96 transition is @xmath79 at @xmath97 .
is not detected in the keck / hires spectrum , nor any other low ionization transitions in the _ hst_/stis spectrum .
however , the system is confirmed by in the _ hst_/fos spectrum . the complex metal - line system at @xmath98 is spread over more than @xmath99 .
@xcite separated it into three systems at @xmath91 ( system a ) , @xmath98 ( system b ) , and @xmath100 ( system c ) .
they suggest that these systems are produced by three different galaxies in a group .
system a , at @xmath91 , has detected , , si0.1emiv @xmath14 , c0.1emiv @xmath15 , and in the _
hst_/stis spectrum .
many other transitions , including o0.1emvi @xmath16 and lyman series lines , were detected in the _
hst_/fos spectrum .
@xmath79 is blended with @xmath32 from the @xmath95 system .
is saturated and covers the entire velocity range between this system and system b. system b , at @xmath97 , produces a partial lyman limit break in the _ hst_/fos spectrum .
it is detected in several transitions of , , , si0.1emiv @xmath14 , c0.1emiv @xmath15 , and .
galactic @xmath101 is blended with the red wing of the profile .
@xmath42 is blended to the red with an unidentified line , probably .
the feature redward of @xmath70 is @xmath96 from the @xmath95 system .
is self - blended , with the @xmath96 from system a superimposed on the @xmath32 from system b. system c , at @xmath102 is unusual in its high velocity relative to the main system , system b. for this system , which would be considered a single - cloud , weak absorber if isolated from the other systems , is detected in , , , , si0.1emiv @xmath14 , and c0.1emiv @xmath15 .
@xmath58 is blended with @xmath103 from system b , but @xmath103 is not detected in a clean part of the spectrum .
again , and various other transitions are detected in the _
hst_/fos spectrum .
the system at @xmath104 is almost certainly intrinsic judging by its broad profile and the `` non - black '' line , evidence for partial covering .
no other metals are detected besides the .
this was also a quasar studied for the quasar absorption line key project @xcite , and a keck / hires spectrum was obtained by @xcite .
particular systems , with optical and uv coverage , were studied in @xcite .
the @xmath106 , 0.5584 and 0.8954 systems were modeled in detail by @xcite . this quasar was observed with _
hst_/stis by churchill et al .
in order to provide constraints on the physical conditions in absorbers .
forest lines appear in _ hst_/stis spectrum quasar spectrum blueward of 2763 , and the forest is covered blueward of 2332 .
we find evidence for six metal - line systems : one strong absorber , two weak absorbers , one weak , and two associated absorbers .
the system plots for these metal - line systems are shown in figures 3338 .
@xcite found a metal - line system at @xmath107 with strong detected .
this system corresponds to a strong absorber detected in a keck / hires spectrum @xcite , which also has detected , , and . in the _ hst_/stis spectrum , we also find @xmath39 , @xmath38 , and c0.1emiv @xmath15 .
there is a @xmath43 detection of @xmath108 at @xmath109 , which coincides with a satellite component of mg0.1emii @xmath13 in the keck / hires spectrum @xcite . for
the multiple - cloud , weak absorber at @xmath110 , c0.1emiv @xmath15 was detected in the _ hst_/stis spectrum , but no low ionization transitions were detected , though @xmath38 , al0.1emiii @xmath37 , and @xmath111 were covered . a possible weak doublet at @xmath112 can not be confirmed , since is not covered in the _ hst_/stis or in the previous _ hst_/fos spectrum .
the single - cloud , weak absorber at @xmath113 has detected , @xmath54 , @xmath79 , and c0.1emiv @xmath15 .
the @xmath114 is affected by a blend .
although there is a detection at the position of @xmath18 , @xcite note that this may be a blend , since it is difficult to explain the strength of this feature in comparison to the weaker mg0.1emii @xmath13 lines .
the system , at @xmath115 , found by @xcite is confirmed here , showing o0.1emvi @xmath16 and in the _
hst_/stis spectrum .
although a feature is detected at the position of @xmath58 , also noted by @xcite , this is inconsistent with the lack of detection of @xmath116 , and so can not be confirmed .
@xcite found a strong associated absorber at @xmath117 in their _
hst_/fos spectrum , for which and were also detected . in the higher resolution _
hst_/stis spectrum , we confirm these detections , but do not find any other associated absorption .
the relatively weak , as compared to o0.1emvi @xmath16 is characteristic of a class of associated absorbers @xcite .
this quasar was studied by the _ hst _
quasar absorption line key project using g190h and g270h spectra obtained with fos @xcite , who found metal - line systems at @xmath119 , 0.7660 , 0.7732 and 0.8553 .
the latter two systems correspond to strong absorbers , modeled by @xcite based on keck / hires data @xcite and the _ hst_/stis spectrum presented here .
another system , at @xmath120 , with only c0.1emiv
@xmath15 and detected , was discussed in @xcite .
@xcite obtained a spectrum of this quasar with _
hst_/stis in order to provide constraints on the physical conditions in the two strong absorbers along the line of sight .
hst_/stis spectrum , the forest is only covered up to a wavelength of @xmath121 , and higher order lyman series lines are not covered .
we find evidence for two strong and two weak systems in the _
hst_/stis spectra .
these systems are shown in figures 3945 .
@xmath38 and @xmath481855 are the only prominent transitions from the possible @xmath119 system that are covered in the _ hst_/stis spectrum .
although there is a detection at the position of @xmath38 , the lack of detected al0.1emiii @xmath37 and the lack of detection of c0.1emiv @xmath15 in the fos spectrum leads us to question the reality of this system .
thus we prefer an identification of the feature at @xmath122 as .
a doublet is found at @xmath123 in the _
hst_/stis spectrum .
it is confirmed by a detection in the g190h _ hst_/fos spectrum @xcite .
the weak feature at the expected position of @xmath124 is unlikely to be because of the absence of the stronger @xmath18 transition in the fos spectrum .
there is a very weak possible doublet at @xmath125 in the _
hst_/stis spectrum , however the alignment is not perfect . also , there is no detected in the fos spectrum to a rest frame equivalent width limit of @xmath126 .
the absorption system at @xmath120 is confirmed by a detection in the fos spectrum , but there are no additional detections in the _ hst_/stis spectrum .
the strong system at @xmath127 has detected @xmath38 , @xmath128 and @xmath124 , @xmath42 , si0.1emiv @xmath14 , and c0.1emiv @xmath15 in the stis e230 m spectrum .
@xmath129 may be detected in the strongest low ionization component , but the spectrum is very noisy in that region . we can not confirm a system at @xmath130 , which was suggested by @xcite , however we note that we also do not detect c0.1emiv @xmath15 to a @xmath43 rest frame equivalent width limit of @xmath131 .
it is still possible that this system exists , but is collisionally ionized , as proposed by @xcite .
the strong system at @xmath132 has detected @xmath18 , @xmath42 , si0.1emiv @xmath14 , c0.1emiv @xmath15 , and .
kinematically , this system is interesting , with satellite components in low ionization gas ( offset @xmath133 from the main absorption ) having corresponding , broader absorption .
this quasar was previously studied through g190h and g270h _ hst_/fos spectra @xcite , but only three metal - line systems were indicated , at @xmath136 , 0.6606 and 1.0022 .
the @xmath137 coincides with a strong absorber @xcite , and it was modeled by @xcite .
the _ hst_/stis spectrum was obtained in a program ( p.i .
churchill ) to facilitate a study of the @xmath137 absorber .
this spectrum covers the forest up to a wavelength of @xmath138 , and does not cover any lines .
besides galactic absorption , in the _
hst_/stis e230 m spectrum , we detect metal - line absorption from only the system at @xmath139 , which corresponds to a full lyman limit break seen in a _
g160l spectrum @xcite .
a system plot is shown in figure 46 . from this system
we find @xmath38 , @xmath124 , and c0.1emiv @xmath15
. is not detected .
in fact , the absorption at the velocity of the low ionization absorption is relatively weak , and this system classifies as `` deficient '' @xcite .
the strongest absorption is @xmath140 redward of the strongest low ionization component .
@xcite note a complex of lines ranging from @xmath141 to @xmath142 .
they found no metals associated with this complex in the low resolution _
hst_/fos spectrum .
we note that with the stis data , it is possible to set a stronger limit , with a rest frame @xmath43 equivalent width limit @xmath143 .
this was one of the _ hst _ qso absorption line key project quasars , and its _ hst_/fos g270 m spectrum was presented in @xcite . in that spectrum ,
three extensive metal - line systems are present at @xmath146 , 0.9908 and 1.0417 . also , there are possible doublets at @xmath147 , 0.6540 , 0.6790 and 0.7796 , and a possible doublet at @xmath148 . in a
keck / hires optical spectrum , strong absorption is detected from the @xmath149 system , and weak absorption from the @xmath150 , 0.9060 and 1.0417 systems @xcite .
weak is also detected at @xmath151 in the optical spectrum @xcite .
the three single - cloud , weak absorbers at @xmath151 , 0.9056 and 0.6534 were modeled by @xcite , the multiple - cloud , weak absorber at @xmath152 by @xcite , and the strong absorber at @xmath153 by @xcite .
two programs contribute different wavelength coverage of this quasar with _
burles ( 1997 , hst proposal i d # 7292 ) observed the quasar in order to measure the primordial d / h ratio .
@xcite provided redward coverage in order to conduct a detailed study of the forest . in the _ hst_/stis spectra , is covered up to @xmath154 , up to @xmath155 , and in the small region up to @xmath156 .
there is detected absorption from the strong absorber , and from the four weak absorbers mentioned above .
there are also several possible systems and a possible system , as well as an interesting `` /-only '' system .
there is a possible intrinsic system and an associated system observed toward this quasar as well .
we report on these systems here , and shown system plots in figures 4756 .
we can only tentatively claim a doublet at @xmath157 .
the @xmath32 and @xmath96 profiles match reasonably well , but @xmath96 may be a bit too strong at some velocities , and there is no coverage of transitions that could confirm this identification
. this could relate to a blend with the @xmath53 from the @xmath158 system ( see below ) .
there is also a probable , narrow doublet at @xmath159 .
the @xmath34 transition is blended with @xmath32 from the @xmath160 system .
if this is real , it is unusual in the weakness of relative to . a weak
, narrow doublet may also be detected .
the _ hst_/stis spectrum shows weak absorption in several low and intermediate transitions related to the @xmath160 single - cloud , weak absorber .
the has two components redward of the dominant low ionization absorption , and is weaker but has similar kinematics .
the single - cloud , weak absorber at @xmath161 has detected , , , and absorption , all centered at the same velocity as the .
@xmath162 is blended with from a system at @xmath163 , and is apparent as a small depression in the red wing of that feature . for another single - cloud
, weak absorber at @xmath164 , , , , , , , , and are detected in the _
hst_/stis spectrum .
the coverage of and is unusual for single - cloud , weak absorbers and is important for considerations of the phase structure of these objects .
there is an interesting system detected in the _
hst_/stis spectrum at @xmath165 .
most of the common metal transitions have good coverage , but only and are detected . and confirm the reality of this system .
the -deficient , strong absorber at @xmath166 has detections in the _
hst_/stis spectrum of the first four lyman series lines , , and a variety of singly to triply ionized transitions , including relatively weak absorption . and are covered , but not detected .
the line is detected , but suffers from a blend to the blue with @xmath167 from the system at @xmath158 .
this system is of interest because the higher ionization transitions appear redward of the lower ionization transitions , indicating a density gradient across the system @xcite .
the @xmath158 multiple - cloud , weak system has a partial lyman limit break detected in a _
hst_/stis g230 m spectrum .
the higher order lyman series lines are detected down to the break in the _
hst_/stis e230 m spectrum , though past @xmath168 they suffer badly from blending with the lyman series lines from the @xmath166 system .
metal lines are detected in , , , , and from each of two subsystems .
the low ionization transitions are centered at different velocities than the high ionization transitions .
@xmath53 is also detected , but may be blended with the possible doublet at @xmath157 . is detected in a low resolution _
hst_/fos spectrum , but is redward of the _ hst_/stis coverage . there is a possible absorber at @xmath169 , however its @xmath50 transition is too strong relative to the @xmath52 transition .
although , , and are detected at this velocity , their relative strengths are also inconsistent , perhaps due to unknown blends . an doublet is detected from a system at @xmath148 , which is slightly redward of @xmath170 .
it is confirmed by a very weak detection , but is badly blended and provides no additional constraint .
this is another example of an associated system ( e.g. @xcite ) . the e230 m observation of this quasar with _ hst_/stis was proposed by burles et al .
( 1997 , hst proposal i d # 7292 ) to facilitate a study of the deuterium to hydrogen ratio in a system at @xmath173 @xcite .
hst_/stis g230 m spectrum was also obtained , covering the lyman series for this system @xcite .
earlier , this quasar was observed as part of the hst quasar absorption line key project . @xcite found metal - line systems at @xmath174 , 1.0323 and 1.0872 . in the _
hst_/stis spectrum , is covered up to @xmath175 , up to @xmath176 , and up to @xmath177 .
all lyman series lines are covered below @xmath178 .
in addition to the @xmath173 system , for which this observation was obtained , we find evidence for four definite and two possible absorbers .
these systems are all shown in figures 5764 .
the @xmath120 lyman limit system has and detected in the _
hst_/stis spectrum .
weak @xmath42 also appear to be detected , consistent with the weak absorption that @xcite found in a hires / keck spectrum .
c0.1emiv @xmath15 and si0.1emiv @xmath14 are covered , but not detected .
an absorber is found at @xmath179 in the stis spectrum , accompanied by detections of , , and .
@xcite also found in a low resolution _
hst_/fos spectrum .
another system is detected at @xmath180 in the _
hst_/stis spectrum . and
are also detected , confirming this doublet .
a stronger absorber is detected at @xmath181 , corresponding to the system found by @xcite . in the _
hst_/stis spectrum the first five lyman series lines are detected , along with , , and o0.1emvi @xmath16 .
there is a feature detected ( just above @xmath10 ) at the position of @xmath50 for a system at @xmath182 with detected and absorption .
the @xmath52 is not detected , even at @xmath183 , so it is uncertain whether there are detected metals for this system . the @xmath184 system , with , , , and detected in the _ hst_/stis spectrum , appears to be an absorber , with weak absorption .
the doublet can not be confirmed because the @xmath185 member is blended with at @xmath186 .
there is an associated system with @xmath187 , detected in , , and .
the profile has a black , saturated region at this redshift .
the @xmath187 component has detections in and .
there is also a non - black component at @xmath188 .
this system detected o0.1emvi @xmath16 , and resembles many other associated systems .
this paper presents a catalog that is useful for investigations of metal - line systems and the forest in eight of the highest s / n - ratio e230 m spectra from the _
hst_/stis archive .
we have identified a total of 56 metal - line systems ( with at least one metal line detected ) toward the eight quasars listed in table [ tab : tab1 ] , as presented in figures 964 . for key transitions , table [ tab : tab4 ]
tabulates in how many of the 56 systems a given transition was covered and in how many it was detected .
we must caution that clearly these numbers apply only to this particular dataset and are highly dependent on the detection limits of the spectra , to the exact wavelength coverage , and to biases toward certain kinds of quasars for the sample . in this sample of eight quasars
, there may be a preference toward strong low ionization absorbers , since many of them were chosen for a program to study the associated with absorption ( see table [ tab : tab1 ] ) . nonetheless , it is useful to note from table [ tab : tab4 ] that for about half of the 56 systems we have coverage of the and doublets , and that in those cases these high ionization transitions were detected @xmath189% of the time .
intermediate ionization transitions such as @xmath48977 and @xmath481207 were also covered about half the time , and were detected in 50%/58% of the sight lines , respectively .
@xmath481335 is also covered for about half of the systems , and was detected 56% of the time , while @xmath481260 was only detected in 36% of the sightlines for which it was covered .
we conclude that almost all 1 @xmath1 @xmath0 @xmath1 1.7 metal - line systems have high ionization absorption detected , but that only half have detected low ionization absorption .
these numbers are consistent with the findings of @xcite , that half of all absorbers at low redshift have detected low ionization absorption ( sometimes weak ) , and that the other half do not .
this is not surprising since the samples used for these studies have considerable overlap .
studies of the statistics of the forest can be significantly affected by metal lines in the forest . with our catalog
we are able to give an indication of the level of contamination by metal lines in the @xmath190 forest .
specifically , we scanned the wavelength regions of our catalog that are in the forest of the quasar , but are redward of the forest .
table [ tab : tab1 ] lists how many lines and how many metal lines were detected in this region for each quasar .
it also shows that 9 - 10 of the features detected in the e230 m quasar spectra are actually galactic absorption .
we conclude that of the @xmath190 forest lines detected toward each of the quasars ( redward of the forest ) , 20 - 33% of the lines in the forest are actually metal lines .
this `` contamination '' is significant enough that it is important to remove it before computing the statistics of the lines , and is clearly itself a complicated function of redshift .
we acknowledge support from nasa under grant nag5 - 6399 nng04ge73 g and by the national science foundation under grant ast 04 - 07138 .
j. r. m. and r. s. l. were partially funded by the nsf research experience for undergraduates ( reu ) program .
dt and dk were supported in part by stsci grants hst - ar-10688 and hst - ar-10288 .
sb was supported by the nsf / reu program in the summer of 2005 .
99 bahcall j. n. , et al .
, 1996 , apj , 457 , 19 bahcall j. n. , et al . , 1993 , apjs , 87 , 1 brown , t. , et al .
2002 , _ hst _ stis data handbook , version 4.0 , ed .
b. mobasher ( baltimore ; stsci ) charlton , j. c. , ding , j. , zonak , s. g. , churchill , c. w. , bond , n. a. , & rigby , j. r. 2003 , apj , 589 , 311 churchill , c. w. , mellon , r. r. , charlton , j. c. , jannuzi , b. t. , kirhakos , s. , steidel , c. c. , & schneider , d. p. 2000 , apjs , 130 , 91 churchill , c. w. , rigby , j. r. , charlton , j. c. , & vogt , s. s. 1999 , apjs , 120 , 51 churchill , c. w. , & vogt , s. s. 2001 , aj , 122 , 679 ding , j. , charlton , j. c. , & churchill , c. w. 2005 , apj , 621 , 615 ding , j. , charlton , j. c. , churchill , c. w. , & palma , c. 2003a , apj , 590 , 746 ding j. , charlton j. c. , bond n. a. , zonak s. g. , churchill c. w. , 2003b , apj , 587 , 551 ganguly r. et al .
, 2007 , in preparation jannuzi b. t. , et al . , 1998 , apjs , 118 , 1 kirkman , d. , tytler , d. , lubin , d. , & charlton , j. 2007 , mnras , 376 , 1227 kirkman , d. , et al . 2001 , apj , 559 , 23 koratkar a. , antonucci r. , goodrich r. , storrs a. , 1998 , apj , 503 , 599 levshakov s. a. , agafonova i. i. , reimers d. , baade r. , 2003 , a&a , 404 , 449 masiero , j. r. , charlton , j. c. , ding , j. , & churchill , c. w. 2005 , apj , 623 , 57 milutinovi n. , rigby j. r. , masiero j. r. , lynch r. s. , palma c. , charlton j. c. , 2006 , apj , 641 , 190 narayanan , a. , charlton , j. c. , masiero , j. r. , & lynch , r. , 2005 , apj , 632 , 92 rao , s. m. , & turnshek , d. a. , 2000 , apjs , 130 , 1 reimers d. , baade r. , quast r. , levshakov s. a. , 2003 , a&a , 410 , 785 savage , b. d. , sembach , k. r. , tripp , t. m. , & richter , p. , 2002 ,
apj , 564 , 631 schneider , d. p. , et al .
1993 , apjs , 87 , 45 tripp , t. m. , savage , b. d. , & jenkins , e. b. 2000 , apj , 534 , l1 verner , d. a. , verner , e. m. , & ferland , g. j. 1996 , adndt , 64 , 1 zonak , s. g. , charlton , j. c. , ding , j. & churchill , c. w. 2004 , apj , 606 , 196 + @xmath191 number of systems in which transition is covered , @xmath192 number of systems in which transition is detected , @xmath193 probability that transition is detected at @xmath194 for system in which it is covered . | we have produced a catalog of line identifications and equivalent width measurements for all absorption features in eight ultraviolet echelle quasar spectra .
these spectra were selected as having the highest signal - to - noise among the _
hst_/stis spectra obtained with the e230 m grating .
we identify 56 metal - line systems toward the eight quasars , and present plots of detected transitions , aligned in velocity - space .
we found that about 1/4 - 1/3 of the features in the forest region , redward of the incidence of the forest , are metal lines .
high ionization transitions are common .
we see both and in 88 90% of the metal - line systems for which the spectra cover the expected wavelength .
is seen in 58% , while low ionization absorption in , , and/or is detected in 50% of the systems for which they are covered .
this catalog will facilitate future studies of the forest and of metal - line systems of various types .
[ firstpage ] intergalactic medium quasars : absorption lines . |
traffic is a realization of an open one - dimensional many - body system . recently ,
popkov and schtz @xcite found that the fundamental diagram determines the phase diagram of such a system , at least for a very simple , yet exactly solvable toy model , the so called asymmetric exclusion process ( asep ) .
in particular , the most important feature that influences the phase diagram is the number of extrema in the fundamental diagram .
this is exactly the theme of this report .
we present an extension of classical , macroscopic ( `` fluid - like '' ) traffic flow models .
usually , it is assumed that the fundamental diagram is a one - hump function , however recent empirical results point to more complicated behaviour .
it is impossible to assign a single flow function @xmath4 to the measured data - points in a certain density range .
therefore , it can be speculated , that this scatter hides a more complicated behaviour of the fundamental diagram in this regime .
we explore two qualitatively different forms of the safe velocity @xmath5 , the velocity to which the flow tends to relax , which leads from the usual one - hump behaviour of the flow density relation to a more complicated function that exhibits , depending on the relaxation parameter , one , two or three humps .
obviously , real drivers may have different @xmath5functions , adding another source of dynamical complexity , which will not be discussed in this paper .
if the behaviour of individual vehicles is not of concern , but the focus is more on aggregated quantities ( like density @xmath1 , mean velocity @xmath6 etc . )
, one often describes the system dynamics by means of macroscopic , fluid - like equations .
the form of these navier - stokes - like equations can be motivated from anticipative behaviour of the drivers .
+ assume there is a safe velocity @xmath7 that only depends on the density @xmath1 .
the driver is expected to adapt the velocity in a way that @xmath6 relaxes on a time scale @xmath8 to this desired velocity corresponding to the density at @xmath9 , @xmath10 if both sides are taylor - expanded to first order one finds @xmath11 inserting @xmath12 @xmath13 abbreviating @xmath14 with @xmath15 the payne equation @xcite is recovered : @xmath16 if one seeks the analogy to the hydrodynamic equations one can identify a `` traffic pressure '' @xmath17
. in this sense traffic follows the equation of state of a perfect gas ( compare to thermodynamics : @xmath18 ) .
+ the above described procedure to motivate fluid - like models can be extended beyond the described model in a straight forward way . if , for example , eq .
( [ ansatz ] ) is expanded to second order , quadratic terms in @xmath8 are neglected , the abbreviation @xmath19 is used and the terms in front of @xmath20 are absorbed in the coupling constant @xmath21 , one finds : @xmath22 the primes in the last equation denote derivatives with respect to the density . since these equations allow infinitely steep velocity changes , we add ( as in the usual macroscopic traffic flow equations @xcite,@xcite ) a diffusive term to smooth out shock fronts : @xmath23 since a vehicle passing through an infinitely steep velocity shock front would suffer an infinite acceleration , we interpret the diffusive ( `` viscosity '' ) term as a result of the finite acceleration capabilities of real world vehicles . our model equations ( [ modeq ] ) extend the equations of the khne - kerner - konhuser ( in the sequel called k@xmath24 model ; @xcite,@xcite ) model by a term coupling to the second derivative of the desired velocity . throughout this study
we use @xmath25 ms@xmath26 , @xmath27 ms@xmath26 and @xmath28 m@xmath29s@xmath26 .
the form of the safe velocity @xmath7 plays an important role in this class of models ( as can be seen , for example , from the linear stability analysis of the @xmath30 model ) .
however , experimentally the relation between this desired velocity and the vehicle density is poorly known .
it is reasonable to assume a maximum at vanishing density and once the vehicle bumpers touch , the velocity will ( hopefully ) be zero .
+ to study the effect of the additional term in the equations of motion we first investigate the case of the conventional safe velocity given by a fermi - function of the form @xcite @xmath31 since @xmath7 is at present stage rather uncertain , we also examine the effects of a more complicated relation between the desired velocity @xmath7 and the density @xmath1 . for this reason we look at a velocity - density relation that has a plateau at intermediate densities , which , in a microscopic interpretation , means that in a certain density regime drivers do not care about the exact distance to the car ahead .
we chose an @xmath7-function of the form @xmath32
\label{vdes_eq}\ ] ] with @xmath33 where @xmath34 is used .
the parameters @xmath35 , @xmath36 and @xmath37 m s@xmath26 are used throughout this study , the corresponding safe velocity and flow are shown in fig .
note that the densities are always normalized with respect to their maximum possible value @xmath38 which is given by the average vehicle length as @xmath39 .
we use a lagrangian particle scheme to solve the navier - stokes - like equations for traffic flow . a particle method similar to the smoothed particle hydrodynamics method
( sph ; @xcite ) has been used previously to simulate traffic flow @xcite , the method we use here , however , differs in the way the density and the derivatives are calculated .
the particles correspond to moving interpolation centers that carry aggregated properties of the vehicle flow , like , for example , the vehicle density @xmath1 .
they are not to be confused with single `` test vehicles '' in the flow , they rather correspond to `` a bulk '' of vehicles .
+ the first step in this procedure is to define , what is meant by the term `` vehicle density '' .
since we assign a number indicating the corresponding vehicle number @xmath40 to each particle @xmath41 with position @xmath42 , the density definition is straight forward , i.e. the number of vehicles per length that can be assigned unambiguously to particle @xmath41 , or @xmath43 once this is done one has to decide in which way spatial derivatives are to be evaluated .
one possibility would be to take finite differences of properties at the particle positions .
however , one has to keep in mind that the particles are not necessarily distributed equidistantly and thus in standard finite differences higher order terms do not automatically cancel out _
exactly_. the introduced errors may be appreciable in the surrounding of a shock and they can trigger numerical instabilities that prevent further integration of the system .
therefore we decided to evaluate first order derivatives as the _ analytical derivatives _ of cubic spline interpolations through the particle positions .
second order derivatives of a variable @xmath44 are evaluated using centered finite differences @xmath45 where @xmath46 and @xmath47 are evaluated by spline interpolation and @xmath48 is an appropriately chosen discretisation length . since we do not evolve the `` weights '' @xmath40 in time , there is no need to handle a continuity equation , the total vehicle number @xmath49 is constant and given as @xmath50 + denoting the left hand side of ( [ modeq ] ) in lagrangian form @xmath51 , we are left with a first order system : @xmath52 this set of equations is integrated forward in time by means of a fourth order accurate runge - kutta integrator with adaptive time step . + the described scheme is able to resolve emerging shock fronts sharply without any spurious oscillations .
an example of such a shock front is shown in fig .
[ shockfront ] for the @xmath30-model .
traffic modelling as well as traffic measurements have a long - standing tradition ( e.g. greenshields @xcite , lighthill and whitham @xcite , richards @xcite , gazis et al .
@xcite , treiterer @xcite , to name just a few ) . in recent years
physicists working in this area have tried to interpret and formulate phenomena encountered in traffic flow in the language of non - linear dynamics ( see , for example , the review of kerner @xcite and many of the references cited therein ) .
the measured real - world data reveal a tremendous amount of different phenomena many of which are also encountered in other non - linear systems .
+ to identify properties of our model equations we apply them to a closed one - lane road loop .
the loop has a length of @xmath53 km and we prepare initial conditions close to a homogeneous state with density @xmath54 ( i.e. same density and velocity everywhere ) .
the system is slightly perturbed by a sinusoidal density perturbation of fixed maximum amplitude @xmath55 and a wavelength equal to the loop length .
the particles are initially distributed equidistantly , the weights @xmath40 are assigned according to ( [ dens ] ) in order to reproduce the desired density distribution , and the velocities corresponding to @xmath5 are used .
all calculations are performed using 500 particles . + it is important to keep in mind that the results are only partly comparable to real world data since the latter may reflect the response of the non - linear system to external perturbations like on - ramps , accidents etc . which are not included in the model .
+ in the following the model parameter @xmath8 , which determines the time scale on which the flow tries to adapt on @xmath7 , is allowed to vary .
this corresponds to a varying acceleration capability of the flow due to a changing vehicle composition ( percentage trucks etc . ) .
this parameter @xmath8 which is typically of the order of seconds controls a wide variety of different dynamical phenomena .
a similar result has been found in @xcite for a microscopic car - following model .
the @xmath56-isocline , i.e. the locus of points in the @xmath57-plane for which @xmath58 , is found from eq .
( [ eq1 ] ) to be the abscissa .
the velocity isocline , i.e. the @xmath59-points where the acceleration vanishes can be inferred from eq .
( [ eq2 ] ) . for the homogeneous and stationary solution one finds the isocline velocity @xmath60 as a function of @xmath1 : @xmath61 fixed - points of the flow , defined as intersections of the @xmath56- and @xmath6-isoclines ,
are thus expected only for densities above @xmath62 , where @xmath60 approaches the abscissa , see fig .
[ force_free_fd ] .
the flow of the homogeneous and stationary solution has no fixed point in a strict sense since @xmath63 becomes extremely small at @xmath64 , but does not vanish exactly .
however , this could easily be changed by choosing another form of @xmath7 .
[ force_free_fd ] shows these `` force free velocities '' @xmath60 ( in the homogeneous and stationary limit ) together with the `` force free fundamental diagrams '' ( for @xmath65 s ) for both investigated forms of @xmath7 .
we expect the fundamental diagrams ( fd ) found from the numerical analysis of the full equation set to be centered around these `` force free fundamental diagrams '' . while for the @xmath7 of the @xmath30-model the fd always ( i.e. for @xmath65 s ) exhibits a simple one - hump behaviour , eq .
( [ vdes_eq ] ) leads to a one , two or three hump structure of the fd depending on the time constant @xmath8 .
that is why a stronger dependence of qualitative features on this constant may be expected for the plateau safe velocity .
note , however , that even with the conventional @xmath7 the `` force free velocity '' @xmath60 exhibits for @xmath66 s two additional extrema at intermediate densities ( up to four with the plateau function ) .
the implications of these additional extrema for the stability of the flow are discussed below .
+ to get a preliminary idea about the stability regimes of the model it is appropriate to perform a _ linear stability analysis_. by inserting ( [ vel_iso ] ) into the equations of motion we obtain equations that formally look like the equations of the @xmath30-model : @xmath67 the role of @xmath7 now being played by @xmath60 .
thus with the appropriate substitution the _ linear stability criterion _ of @xcite can be used : @xmath68 \frac { c_0}{\rho}. \label{stab_crit}\ ] ] thus we expect the flow to be linearly unstable in density regimes where the decline of @xmath60 with @xmath1 is steeper than a given threshold .
specifically , extrema of the @xmath69 are ( to linear order ) stable and we therefore expect _ stable density regions embedded in unstable regimes_. the previous analytical considerations give a rough idea of what to expect , for a more complete analysis , however , we have to resort to a numerical treatment of the full equation set . in order to be able to distinguish the effects resulting from the additional term in the equations of motion from those coming from the form of @xmath7 we treat two cases separately : in the first case the conventional form of @xmath7 is used and in the second the effects due to a plateau in @xmath7 @xcite are investigated . to obtain a * fundamental diagram * ( fd ) comparable to measurements we chose a fixed site on our road loop .
we determine averages over one minute in the following way : @xmath70 and @xmath71 , where @xmath72 is the number of particles that have passed the reference point within the last minute . the thus calculated fd ( fig .
[ fd_alltau ] , left panel ) is , as expected , close to a superposition of the `` force - free fds '' , for different values of @xmath8 see fig .
[ force_free_fd ] , left panel .
as in real - world traffic data in the higher density regimes the flow is not an unambiguous function of @xmath1 , but rather covers a surface given by the range of @xmath8 in the measured data .
note that many data points in the unstable regime ( see below ) exhibit substantially higher flows than expected from the `` force - free fds '' ( see fig .
[ force_free_fd ] ) .
+ in certain density ranges the model shows * instability * with respect to jam formation from an initial slight perturbation .
in this regime the initial perturbation of the homogeneous state grows and finally leads to a breakdown of the flow into a backward moving jam ( kerner refers to this state , where vehicles come in an extended region to a stop , as `` wide jam '' ( wj ) contrary to a `` narrow jam '' ( nj ) which basically consists only of its upstream and downstream fronts and vehicles do not , on average , come to a stop ; @xcite ) .
this phenomenon , widely known as `` jam out of nowhere '' , is reproducible with several traffic flow models ( e.g. @xcite ) .
an example of a spontaneously forming wj accompanied by two njs is given in fig .
[ widejam ] , left panel , for an initial density of @xmath73 .
it is interesting to note that the initial perturbation remains present in the system for approximately 15 minutes without noticeably growing in amplitude before the flow breaks down . as in reality
the inflow front of the wj is much steeper than the outflow front of the jam .
note the similarity with the jam formation process within the k@xmath24-model @xcite .
+ to give a global idea in which density regimes congestion phenomena occur we show in fig .
[ velvar ] , left panel , the velocity variance @xmath74 for given initial densities .
the system is allowed to evolve from its initial state until @xmath75 converges . if @xmath75 has not converged after a very long time @xmath76 (= 10000 s ) it is assumed that no stationary state ( @xmath77 const ) can be reached and @xmath75 is taken at @xmath76 .
@xmath49 denotes the particle number and @xmath78 the average particle velocity in the system . for low values of @xmath8 ( 1 and 5 s ) the system shows spontaneous jam formation in a coherent density regime from @xmath79 to @xmath80 , comparable to measured data . for @xmath66 s a stable regime at intermediate densities surrounded by unstable density regimes
is encountered .
this region corresponds to the two close extrema seen in fig .
[ force_free_fd ] , first panel . +
another widespread phenomenon is the formation of several jams following each other , so - called * stop - and - go - waves*. this phenomenon is also a solution of our model equations , see fig .
[ stop_and_go ] , left panel .
the emerging pattern of very sharply localized perturbations is found in empirical traffic data as well ( see fig .
14 , detector d7 in @xcite ) . +
a very interesting phenomenon happens towards the upper end of the instability range ( @xmath81 ) .
after the initial perturbation has remained present in the system for more than 20 minutes without growing substantially in amplitude , see fig .
[ plateau1 ] , suddenly a sharp velocity spike appears at @xmath82 s that broadens in the further evolution until the system has separated into two phases : a totally queued phase , where the velocity vanishes on a distance of several kilometers , and a homogeneous high velocity phase , both separated by a shock - like transition .
we refer to these states with homogeneous velocity plateaus separated by shock fronts as * mesa states*. the numerically determined * fundamental diagrams * for the case with plateau is shown in fig .
[ fd_alltau ] , right panel .
the additional extrema expected from the `` force - free velocity '' @xmath60 are visible in the data points .
we therefore conclude that _ if a pronounced plateau in @xmath7 really does exist , additional extrema should appear in the measured fundamental diagrams , at least for flows with poor acceleration capabilities , i.e. large @xmath8s_. + also with the plateau function the system shows * spontaneous jam appearance*. the formation of an isolated , stable wj is displayed in fig .
refwidejam , right panel . with a change in the parameter @xmath8 ( 10 s rather than 5 s as in fig . [ widejam ] )
one finds a more complicated pattern with one wj that coexists for a long time with constantly emerging and disappearing njs , see fig .
[ jamsync ] .
+ the global stability properties for the case with plateau are shown in fig .
[ velvar ] , right panel . as expected from the linear stability analysis
( [ stab_crit ] ) and fig .
[ force_free_fd ] , right panels ) we find alternating regimes of stability and instability rather than one coherent density range where the flow is prone to instability . for low ( @xmath83 ) and very high density ( @xmath84 ) , initial perturbations decrease in amplitude , i.e. the system relaxes towards the homogeneous state . in between these density perturbations
may grow and lead to spontaneous structure formation of the flow .
the stable regions within unstable flow are found around densities for which @xmath85 .
this is displayed for two values of @xmath8 in fig .
[ velvar_comp ] .
+ the * accelerations * in the model were never found to exceed @xmath86 ms@xmath29 for negative and @xmath87 ms@xmath29 for positive signs and thus agree with accelerations from real - world traffic data ( for both forms of @xmath7 ) . for reasons of illustration fig .
[ acc ] displays velocities and the corresponding accelerations at one time slice of a simulation ( @xmath88 ms@xmath89 s ) for @xmath7 according to eq .
( [ vdes_eq ] ) .
also the plateau function allows for * stop - and - go - waves * , see fig . [ stop_and_go ] , right panel .
the shown evolution process is close to what kerner @xcite describes as general features of stop - and - go - waves : initiated by a local phase transition from free to synchronized flow , numerous well localized njs emerge , move through the flow and begin to grow .
one part of the njs propagates in the downstream direction ( see e.g. the perturbations located at @xmath90 km at @xmath91 s ) while the rest ( at @xmath92 s at @xmath93 km ) move upstream .
once the first wj has formed after approximately 500 s the njs start to merge with it .
this nj - wj merger process continues until a stationary pattern of three wjs has formed ( at around 1000 s ; not shown ) which moves with constant velocity in upstream direction .
the distance scale of the downstream fronts of these self - formed wjs is in excellent agreement with the experimental value of 2.5 - 5 km @xcite .
+ we found for the conventional form of @xmath7 a separation into different homogeneous velocity phases that we called * mesa states*. this feature is also present if the plateau function is used . in fig .
[ plateau1 ] , right panel , the initial perturbation organizes itself into different platoons of homogeneous velocities .
these platoons are separated by sharp , shock - like transitions and form a stationary pattern that moves along the loop without changing in shape . + the relaxation term in eq .
( [ modeq ] ) plays a crucial role for the stabilisation of this pattern . if , for example , the relaxation time @xmath8 is increased ( see fig .
[ plateau2 ] ) and thus the importance of the relaxation term is reduced , the system is not able to stabilize the velocity plateau .
it seems to be aware of these states , but it is always heavily disturbed and never able to reach a stationary state .
again , the composition of the traffic flow plays via @xmath8 _ the _ crucial role for the emerging phenomena .
starting from the assumption that a safe velocity @xmath5 exists towards which drivers want to relax by anticipating the density ahead of them , we motivate a set of equations for the temporal evolution of the mean flow velocity .
the resulting partial differential equations possess a navier - stokes - like form , they extend the well - known macroscopic traffic flow equations of khne , kerner and konhuser by an additional term proportional to the second derivative of @xmath5 .
motivated by recent empirical results , we explore , in addition to the new equation set , also the effects of a @xmath7-function that exhibits a plateau at intermediate densities .
the results are compared to the use of a conventional form of @xmath7 .
+ these fluid - like equations are solved using a lagrangian particle scheme , that formulates density in terms of particle properties and evaluates first order derivatives analytically by means of cubic spline interpolation and second order derivatives by equidistant finite differencing of the splined quantities .
the continuity equation is fulfilled automatically by construction .
this method is able to follow the evolution of the ( in some ranges physically unstable ) traffic flow in a numerically stable way and to resolve emerging shock - fronts accurately without any spurious oscillations .
+ the presented model shows for both investigated forms of @xmath7 a large variety of phenomena that are well - known from real - world traffic data .
for example , traffic flow is found to be unstable with respect to jam formation initiated by a subtle perturbation around the homogeneous state . as in reality
, stable backward - moving _ wide jams _ as well as sharply localized _ narrow jams _ form .
these latter ones move through the flow without leading to a full breakdown until they merge and form wide jams .
the distance - scale of the downstream fronts of these self - formed wide jams is in excellent agreement with the empirical values .
the encountered accelerations are in very good agreement with measured values . for the @xmath7 with plateau
we also find states where a stable wide jam coexists with narrow jams that keep emerging and disappearing without ever leading to a breakdown of the flow , properties that are usually attributed to the elusive state of `` synchronized flow '' .
another , striking phenomena is encountered that we call the `` mesa - effect '' : the flow may organize into a state , where platoons of high and low velocity follow each other , separated only by a very sharp , shock - like transition region .
this pattern is found to be stationary , i.e. it moves forward without changing shape .
one may speculate , that these mesa states are related to the minimum flow phase found in the work using the asep as a model for traffic flow @xcite .
+ in other regions of parameter space the flow is never able to settle into a stationary state . here
wide jams and a multitude of emerging moving or disappearing narrow jams may coexist for a very long time .
again , it may be presumeded that in these cases the system displays deterministic chaos , however , we did not check this beyond any doubt .
+ the basic effect of the new interaction term is to make the `` force - free velocity '' , which essentially determines the shape of the fundamental diagram , sensitive to the relaxation parameter @xmath8 . for large values of @xmath8 additional extrema in the `` force - free flows '' are introduced and a stability analysis shows that that the flows are stable against perturbations in the vicinity of these extrema .
this leads to the emergence of alternating regimes of stability and instability , the details of which depend on the shape of @xmath7 .
we find that if a pronounced plateau in @xmath7 really does exist , it should appear in the measured fundamental diagrams , at least for flows with poor acceleration capabilities , i.e. large @xmath8 s .
+ the crucial parameter besides density which determines the dynamic evolution of the flow and all the related phenomena is the relaxation time @xmath8 .
since this parameter governs the time scale on which the flow tries to adapt to the desired velocity @xmath7 , we may interpret it as a measure for the flow composition ( fraction of trucks etc . ) .
it is this composition that determines whether / which structure formation takes place , whether the system relaxes into a homogeneous state , forms isolated wide jams or a multitude of interacting narrow jams . to conclude
, this work shows that a surprising richness of phenomena is encountered if one allows for a slight change of the underlying traffic flow equations .
further work is needed in order to extend the qualitative description undertaken in this work and to find more quantitative relationships between the traffic flow models and reality . | based on the assumption of a safe velocity @xmath0 depending on the vehicle density @xmath1 a macroscopic model for traffic flow is presented that extends the model of the khne - kerner - konhuser by an interaction term containing the second derivative of @xmath2 .
we explore two qualitatively different forms of @xmath3 : a conventional , fermi - type function and , motivated by recent experimental findings , a function that exhibits a plateau at intermediate densities , i.e. in this density regime the exact distance to the car ahead is only of minor importance . to solve the fluid - like equations a lagrangian particle scheme is developed .
+ the suggested model shows a much richer dynamical behaviour than the usual fluid - like models .
a large variety of encountered effects is known from traffic observations many of which are usually assigned to the elusive state of `` synchronized flow '' .
furthermore , the model displays alternating regimes of stability and instability at intermediate densities , it can explain data scatter in the fundamental diagram and complicated jam patterns . within this model , a consistent interpretation of the emergence of very different traffic phenomena
is offered : they are determined by the velocity relaxation time , i.e. the time needed to relax towards @xmath2 .
this relaxation time is a measure of the average acceleration capability and can be attributed to the composition ( e.g. the percentage of trucks ) of the traffic flow . |
nearly 30 years ago , padamanabhan @xcite performed a simple calculation suggesting that quantum gravitational fluctuations place a lower bound on distance measurements .
he considered fluctuations of the conformal factor @xmath0 in metrics of the form @xmath1 while keeping the background metric @xmath2 classical . crudely speaking
, padmanabhan argued that the conformal factor @xmath0 has a green s function that diverges as @xmath3 , in such a way that @xmath4 remains finite in the coincidence limit .
this calculation was part of a larger approach to quantum gravity and quantum cosmology in which the conformal factor was treated as a dynamical field to be quantized , while the rest of the metric was treated as a classical field as in standard qft .
this approach sidesteps some of the thornier conceptual problems associated with quantizing the metric , since conformal fluctuations preserve the causal structure of spacetime .
however , this calculation is almost certainly wrong . in pure einstein - hilbert gravity ,
the conformal factor is not a dynamical degree of freedom @xcite .
this is most clearly seen using the york decomposition of symmetric tensors @xcite , in which the conformal factor is determined by a constraint equation similar to the gauss law constraint in electrodynamics . to see where the argument went wrong
, we must examine the path - integral approach taken by padmanabhan and narlikar @xcite .
the classical action and path integral are @xmath5 = \frac{1}{16 \pi g } \int \rmd^4 x \sqrt{-g}(r-2\lambda ) \label{ehaction}\\ z = \int [ \mathcal{d } g ] \textrm{exp}\left\{\rmi s[g]\right\ } \label{pathintegral}\end{aligned}\ ] ] in terms of the conformal factor and background metric , the action becomes @xmath6 = \frac{1}{16\pi g}\int \rmd^4 x \sqrt{-\bar{g } } \left[\bar{r}(1+\phi(x))^2 - 2\lambda ( 1+\phi(x))^4 - 6\phi^i\phi_i\right ] \label{ehconformalaction}\ ] ] from here , the calculation proceeds in a straightforward manner .
consider the expectation value of the interval in a ( minkowski ) vacuum state @xmath7 : @xmath8 however , @xmath9 evaluated at a single event diverges .
using covariant point - splitting , we instead evaluate the interval between _ two events _ @xmath10 and @xmath11 , in the limit that @xmath12 . with the notation @xmath13 , we examine @xmath14 with @xmath15 , the action is just the action for a massless scalar field , albeit with a negative sign . ] , @xmath16=-\frac{1}{2 l_p^2}\int \phi^i \phi_i \rmd^4x$ ] . the green s function is @xmath17 and so the interval becomes @xmath18 in other words , quantum fluctuations produce a `` ground state length '' just as a harmonic oscillator has a ground state energy .
note that the path integral approach taken here obscures the fact that the conformal factor is not a true dynamical field subject to quantum fluctuations .
the source of this confusion is the apparent kinetic term in the action , which justifies all subsequent steps leading to .
however , in the hamiltonian framework , the trace part of the metric perturbations does not have a canonically conjugate momentum , and a true kinetic term for the conformal factor should not appear in the action .
the explanation for the offending term is hidden in the measure of and was finally resolved by mazur and mottola @xcite . to identify the correct measure , they first decomposed the space of metric perturbations into diffeomorphisms and physical fluctuations .
the remaining physical subspace was further decomposed into constrained ( conformal ) and dynamical ( transverse - traceless ) degrees of freedom .
seen in this light , amounts to a change of coordinates in the space of metrics , which introduces a non - trivial jacobian in the measure .
a field redefinition of the conformal factor then turns the apparent kinetic term in into a potential term , confirming the result that the conformal modes are non - propagating constrained modes .
while the conformal factor is non - propagating in pure einstein - hilbert gravity , the classical constraints that fix the conformal part of the metric fluctuations in terms of matter sources can not be maintained upon quantization @xcite .
the trace anomaly of matter coupled to gravity induces an effective action for the conformal factor that gives rise to non - trivial dynamics @xcite . in other words ,
the conformal factor is promoted to a dynamical field when gravity is coupled to quantized matter .
thus we can revisit padmanabhan s calculation in light of this dynamical model of the conformal factor .
we begin by summarizing the basic results of antoniadis , mazur and mottola @xcite .
the effective action of the conformal factor becomes local in the conformal parameterization @xmath19 where @xmath20 is a fiducial metric .
the total effective action is @xmath21 where @xmath22 is the einstein - hilbert action evaluated at @xmath23 , @xmath24 is the action for matter fields , and @xmath25 is the trace anomaly - induced effective action @xcite @xmath26 = \int \rmd^4x \sqrt{-\bar{g } } \left[2b^\prime \sigma \bar{\delta}_4 \sigma + b^\prime \left(\bar{e } - \frac{2}{3 } \bar{\box } \bar{r}\right)\sigma + b\bar{f}\sigma \right ] .
\label{sanom}\ ] ] here , @xmath27 is the conformally invariant fourth - order operator @xmath28 and @xmath29 are the square of the weyl tensor and the gauss - bonnet integrand , respectively . the coupling constants @xmath30 and @xmath31 depend on the matter content of the theory @xcite : @xmath32 where @xmath33 , @xmath34 and @xmath35 are the numbers of scalar , weyl fermion , and vector fields .
the spin-0 and ghost contributions are included in the -8 and -28 factors , while @xmath36 and @xmath37 count the contributions from the spin-2 metric fields .
because the values of these gravitational contributions , as well as contributions beyond the standard model , remain open questions , @xmath38 will be treated as a free parameter .
the total trace anomaly of the full theory described by must vanish @xcite .
the absence of this anomaly requires that the vacuum is a conformal fixed point at which the @xmath39 functions of all couplings must vanish .
the physical metric then acquires an anomalous scaling dimension @xmath40 where @xmath41 is determined by the @xmath39 function for the einstein - hilbert action @xcite , @xmath42 from here we can follow padmanabhan s prescription . looking only at conformal fluctuations and choosing a minkowski fiducial metric @xmath7 , the action reduces to @xmath43 = -\frac{q^2}{(4\pi)^2}\int \rmd^4x \sigma \bar{\box}^2 \sigma + \frac{1}{8\pi g}\int \rmd^4x \left[3e^{2\alpha \sigma}\left(\partial_a \sigma\right)^2 - \lambda e^{4 \alpha \sigma}\right ] .
\label{conformaleffective}\ ] ] the action simplifies again by invoking the translational invariance of the measure and shifting @xmath44 by a constant @xmath45 @xcite . in the limit @xmath46 , the final terms drop out , leaving only the free quartic action .
the propagator for this fourth - order kinetic term is @xmath47 in momentum space , which is just a logarithm in coordinate space : @xmath48 , \label{propagator}\ ] ] where @xmath49 is an infrared cutoff . now
the expectation value of the interval becomes @xmath50^{2-\alpha^2/2q^2 } , \label{metricexpectation}\end{aligned}\ ] ] where the first line makes use of covariant point - splitting , and normal ordering is used on each operator @xmath51 individually .
the second equality in is a standard field theory result that makes use of the baker - campbell - hausdorff relation @xcite and minkowski space @xcite confirm both of these requirements . ] .
this yields the interesting result that the scaling depends on the matter content : the distance approaches zero for all values @xmath52 or @xmath53 . from ,
positive values of @xmath54 are excluded at the conformal fixed point .
the interval is constant at the critical point @xmath55 , and @xmath56 gives the nonsensical result that distances diverge in the limit @xmath57 . for large @xmath38 ,
the interval scales as @xmath58 , and classical scaling is recovered in the limit @xmath59 .
it follows from that @xmath60 for normal matter ; however , it is worth noting that some models of conformal supergravity contribute negatively to @xmath38 @xcite .
calculations of the one - loop contributions from einstein gravity place it at @xmath61 @xcite .
together , the standard model particle content ( @xmath62 and @xmath63 ) and one - loop gravitational contributions give a value @xmath64 the greatest uncertainty in the value of @xmath38 comes from the gravitational contributions , and a precise theoretical prediction for @xmath38 remains an open problem .
recent attempts to place observational limits on @xmath38 using wmap data claim to limit @xmath38 to the range @xmath65 @xcite .
thus a more complete treatment of conformal fluctuations using the trace anomaly - induced effective action do not place a lower bound on the distance between two points .
of course this result should be viewed with some skepticism . in particular , the spin-2 metric fluctuations are expected to become important around the planck scale but have been frozen out in this approach .
additionally , the transition from einstein gravity to the conformally invariant phase described by is poorly understood , and more research is need to determine the scales at which the effective action becomes significant .
i would like to thank steve carlip for comments on early versions of this paper .
this work was supported in part by the department of energy grant de - fg02 - 91er40674 .
99 padmanabhan t 1984 planck length as the lower bound to all physical length scales _ gen .
* 17 * 215 kuchar k 1970 ground state functional of the linearized gravitational field _ j. math .
phys . _ * 11 * 3322 + york j w 1972 role of conformal three - geometry in the dynamics of gravitation _ phys .
_ * 28 * 1082 + fradkin
e s and vilkovisky g 1973 s matrix for gravitational field .
ii . local measure ; general relations ; elements of renormalization theory _ phys .
_ d * 8 * 4241 + hartle j b 1984 ground - state wave function of linearized gravity _ phys .
_ d * 29 * 2730 antoniadis i , mazur p o and mottola e 1992 conformal symmetry and central charges in 4 dimensions _ nuc phys _
b * 388 * 627 ( _ preprint _ hep - th/9205015 ) antoniadis i and mottola e 1992 four - dimensional quantum gravity in the conformal sector _ phys .
_ d * 45 * 2013 christensen s m and duff m j 1980 quantizing gravity with a cosmological constant _ nucl .
_ b * 170 * 480 + fradkin e s and tseytlin a a 1984 on the new definition of off - shell effective action _ nucl . phys . _ * 234 * 509 antoniadis i , mazur p o and mottola e 1997 physical states of the quantum conformal factor _ phys .
_ d * 55 * 4770 ( _ preprint _ hep - th/9509169 ) + hamada k j 2009 conformal field theory on @xmath66 from quantized gravity _ int .
phys . _ a * 24 * 3073 ( _ preprint _
hep - th/0811.1647 ) hamada k j 2011 background free quantum gravity based on conformal gravity and conformal field theory on @xmath67 _ preprint _ hep - th/1109.6109 antoniadis i , mazur p o and mottola e 2007 cosmological dark energy : prospects for a dynamical theory
_ new j. phys . _ * 9 * 11 ( _ preprint _ gr - qc/0612068 ) + bili n , guberina b , horvat r , nikoli and tefani 2007 on cosmological implications of gravitational trace anomaly _ phys lett .
_ b * 657 * 232 ( _ preprint _ hep - th/0707.3830 ) | this paper corrects an earlier work suggesting that the quantum expectation value of the proper length is bounded from below by the planck length .
the original calculation examined fluctuations of the conformal factor of einstein - hilbert gravity . however , in einstein - hilbert gravity , the conformal factor is not a dynamical field subject to fluctuations .
this paper performs the same calculation using the trace anomaly - induced effective action for the conformal factor and finds that , while conformal fluctuations modify the short - distance behavior of the interval , the interval still approaches zero in the coincidence limit . |
in a recent article @xcite , we have derived exact results for both the temperature @xmath0 and external ordering field @xmath1 behavior of the order parameter profile and the corresponding response functions local and total susceptibilities within the three - dimensional continuum mean - field ginzburg - landau ising type model of a simple fluid or binary liquid mixture for a system with a film geometry @xmath2 . in the current article
we extend them to derive exact results for the thermodynamic casimir force within the same model .
we concentrate in the region of the parametric space in @xmath3 plane close to the critical point of the fluid or close to the demixing point of the binary liquid mixture .
we recall that for classical fluids in the case of a simple fluid or for binary liquid mixtures the wall generically prefers one of the fluid phases or one of the components . because of that in the current article we study the case when the bounding surfaces of the system strongly prefer one of the phases of the system . since in such systems
one observes also the phenomena of the capillary condensation close below the critical point for small negative values of the ordering field @xmath4 , we also study the behavior of the force between the confining surfaces of the system in that parametric region . here
@xmath5 , the field @xmath1 is measured in units of bohr magneton @xmath6 , and @xmath7 is some characteristic microscopic length , say , the average distance between the constituents of the fluid .
let us recall that the model we are going to consider is a standard model within which one studies phenomena like critical adsorption @xcite , wetting or drying @xcite , surface phenomena @xcite , capillary condensation @xcite , localization - delocalization phase transition @xcite , finite - size behavior of thin films @xcite , the thermodynamic casimir effect @xcite , etc .
the results of the model have been also used to calculate the casimir forces in systems with chemically or topographically patterned substrates , as well as , coupled with the derjaguin approximation , for studies on interactions of colloids see , e.g. , the review @xcite and the literature cited therein . until very recently ,
i.e. before ref .
@xcite , the results for the case @xmath8 were derived analytically @xcite while the @xmath1-dependence was studied numerically either at the bulk critical point of the system @xmath9 , or along some specific isotherms see , e.g. , @xcite . in the current article we are going to improve this situation with respect to the casimir force by deriving exact analytical results for it in the @xmath3 plane . in 1948
@xcite , after a discussion with niels bohr @xcite , the dutch physicist h. b. g. casimir realized that the zero - point fluctuations of the electromagnetic field in vacuum lead to a force of attraction between two perfectly conducting parallel plates and calculated this force . in 1978
fisher and de gennes @xcite pointed out that a very similar effect exists in fluids with the fluctuating field being the field of its order parameter , in which the interactions in the system are mediated not by photons but by different type of massless excitations such as critical fluctuations or goldstone bosons ( spin waves ) .
nowadays one usually terms the corresponding casimir effect the critical or the thermodynamic casimir effect @xcite .
currently the casimir , and casimir - like , effects are object of studies in quantum electrodynamics , quantum chromodynamics , cosmology , condensed matter physics , biology and , some elements of it , in nano - technology .
the interested reader can consult the existing impressive number of reviews on the subject @xcite .
so far the critical casimir effect has enjoyed only two general reviews @xcite and few concerning specific aspects of it @xcite .
the critical casimir effect has been already directly observed , utilizing light scattering measurements , in the interaction of a colloid spherical particle with a plate @xcite both of which are immersed in a critical binary liquid mixture . very recently
the nonadditivity of critical casimir forces has been experimentally demonstrated in @xcite .
indirectly , as a balancing force that determines the thickness of a wetting film in the vicinity of its bulk critical point the casimir force has been also studied in @xmath10he @xcite , @xcite , as well as in @xmath11he@xmath10he mixtures @xcite . in @xcite and @xcite
measurements of the casimir force in thin wetting films of binary liquid mixture are also performed .
the studies in the field have also enjoined a considerable theoretical attention .
reviews on the corresponding results can be found in @xcite . before turning exclusively to the behavior of the casimir force ,
let us briefly remind some basic facts of the theory of critical phenomena . in the vicinity of the bulk
critical point @xmath12 the bulk correlation length of the order parameter @xmath13 becomes large , and theoretically diverges : @xmath14 , @xmath15 , and @xmath16 , where @xmath17 and @xmath18 are the usual critical exponents and @xmath19 and @xmath20 are the corresponding nonuniversal amplitudes of the correlation length along the @xmath21 and @xmath1 axes .
if in a finite system @xmath13 becomes comparable to @xmath22 , the thermodynamic functions describing its behavior depend on the ratio @xmath23 and take scaling forms given by the finite - size scaling theory .
for such a system the finite - size scaling theory @xcite predicts : @xmath24 for the casimir force @xmath25 @xmath24 for the order parameter profile @xmath26 where @xmath27 , @xmath28 . in eqs .
( [ cas ] ) and ( [ mfss ] ) , @xmath29 is the critical exponent for the order parameter , @xmath30 is the dimension of the system , @xmath31 and @xmath32 are nonuniversal metric factors that can be fixed , for a given system , by taking them to be , e.g. , @xmath33^{1/\nu}$ ] , and @xmath34^{\delta/\nu}$ ] .
here , as in @xcite , we consider a critical system of ising type in a film geometry @xmath35 , where @xmath22 is supposed to be along @xmath36 axis , described by the minimizers of the standard @xmath37 ginzburg - landau functional @xmath38 = \int_0^l { \cal l}(\phi,\phi ' ) dz,\ ] ] where @xmath39 here @xmath22 is the film thickness , @xmath40 is the order parameter assumed to depend on the perpendicular position @xmath41 only , @xmath42 is the bare reduced temperature , @xmath1 is the external ordering field , @xmath43 is the bare coupling constant and the primes indicate differentiation with respect to the variable @xmath36 .
the thermodynamic casimir force is the excess pressure over the bulk one acting on the boundaries of the system which is due to the finite size of the system . to derive this excess pressure
there are several ways but probably the most straightforward one is to apply the corresponding mathematical results of the variational calculus . for example , following gelfand and fomin @xcite it is easy to show that the functional derivative of @xmath44 with respect to the independent variable @xmath36 at @xmath45 is @xmath46 having in mind eq .
( [ fdefising ] ) , one derives explicitly @xmath47 this derivative has the meaning of a force acting on the surface of the system at @xmath45 and , since @xmath44 is normalized per unit area , it has a meaning of a pressure acting on that surface .
that is why , the notation @xmath48 is used .
the above is actually the procedure used in @xcite where the authors perform the corresponding variational calculations on their own .
another common way to proceed is to use the apparatus based on the stress tensor operator ( see , e.g. , @xcite , @xcite and @xcite ) @xmath49 it is elementary to check that the expression in the parentheses in the right - hand - side of eq .
[ dz ] coincides with @xmath50 component of the stress tensor . in @xcite , we have shown that this expression is a first integral of the considered system , and therefore @xmath50 and @xmath48 do not depend on the coordinate @xmath36 at which they are calculated . in the bulk system , within the mean - field theory
the gradient term in @xmath51 is absent and instead of @xmath52 one obtains @xmath53 where @xmath54 is the order parameter of the bulk system .
clearly , @xmath54 is determined by the cubic equation @xmath55+h=0 $ ] , @xmath54 being such that @xmath56 attains its minimum .
now , one can immediately determine the casimir force as @xmath57 when @xmath58 the excess pressure will be inward of the system that corresponds to an _ attraction _ of the surfaces of the system towards each other and to a _ repulsion _ if @xmath59 . in the light of the above it
is evident that once the order parameter profile @xmath60 is known in analytic form for given values of the parameters @xmath61 and @xmath1 , then the corresponding casimir force is determined exactly .
it is noteworthy that the above expressions do not depend on the specific choice of the boundary conditions . in the current article
we specialize to the so - called @xmath62 boundary conditions under which one requires that @xmath63 .
the exact solution for @xmath64 for this case has been determined in @xcite . in what follows we will study the properties of the force @xmath65 using this exact solution . .
we recall that at this point the susceptibility of the finite system diverges .
the inset on the right shows the pre - capillary - condensation curve , determined in @xcite , where above @xmath66 and below @xmath67 the jump of the order parameter at the middle of the system is from a less dense gas to a more dense one . for @xmath68
the system jumps from a
gas to a liquid state . ]
since the thermodynamic casimir force is normally presented in terms of the scaling variables @xmath69 @xmath70 in the remainder we are going to use such variables as the basic parameters determining the behavior of the force . in the above we have taken into account that for the model considered here @xmath71 @xcite , @xmath72 and @xmath73 .
the phase diagram of this model has been studied in details in @xcite see there figure 3 and the text around it . here , for the convenience of the reader , it is depicted in figure [ fig : pd ] in terms of the scaling variables @xmath74 and @xmath75 . in terms of the scaling variables given in equations ( [ isvarxt ] ) and ( [ isvarxh ] ) ,
the value @xmath48 of the first integral , see eq .
( [ dz ] ) , becomes @xmath76 where the constant @xmath77 is @xmath78 here @xmath79 is the scaling function of the order parameter @xmath60 , @xmath80 and hereafter the prime means differentiation with respect to the variable @xmath81 $ ] .
similarly , for the bulk system , see eq .
( [ pb ] ) , one has @xmath82 where @xmath83 from eqs .
( [ fix ] ) and ( [ pbx ] ) for the casimir force ( [ casimir ] ) one obtains @xmath84 where its scaling function @xmath85 is @xmath86 given @xmath74 and @xmath87 , the determination of @xmath88 is evident , while @xmath77 is given by the expression @xmath89 see eq .
( 3.15 ) in @xcite . as shown in @xcite , @xmath90 is to be determined from @xmath91 so that it gives rize to a continuous order parameter profile in the interval @xmath92 , and satisfies the condition @xmath93 in eq .
( [ treq ] ) @xmath94 is the weierstrass elliptic function whose invariants @xmath95 and @xmath96 are given by the expressions @xmath97 @xmath98.\ ] ] thus , in order to determine @xmath77 for the regarded @xmath62 boundary conditions at given values of the parameters @xmath74 and @xmath87 , one should find all the solutions @xmath90 of the transcendental equation ( [ treq ] ) which meet the above requirements .
if there is more than one such solution @xmath99 , as explained in detail in @xcite , we take that one which leads to an order parameter profile that corresponds to the minimum of the energy functional ( [ fviafising ] ) @xmath100 where @xmath101 the precise mathematical procedure how this can be achieved , despite the divergence of the energy ( see eq .
( 3.27 ) in @xcite and the text around it ) , is also explained in details in @xcite .
let us note that @xmath90 has a clear physical meaning
it is the value of the scaling function of the order parameter profile at the middle of the system , i.e. , @xmath102 . from eqs .
[ pbx1 ] and [ intconst ] , once @xmath103 and @xmath90 are determined , the scaling function of the casimir force takes the form @xmath104 when @xmath8 , i.e. @xmath105 , the behavior of the casimir force under ( + , + ) boundary conditions has been analytically studied in @xcite and @xcite . in @xcite the value of the so - called casimir amplitude , i.e. , the result for @xmath106 is obtained , while in @xcite the behavior of the force as a function of @xmath74 has been studied . when @xmath107 the behavior of the casimir force has been studied only numerically . in refs . @xcite and
@xcite it has been obtained only for @xmath9 for some chosen values of @xmath87 .
below we present its behavior as a function of both @xmath74 and @xmath87 in @xmath108 plane by evaluating numerically the analytical expressions given above . and @xmath87 .
, title="fig:",width=422 ] and @xmath87 . ,
title="fig:",width=422 ] 0.775 , or @xmath87 , as a function of @xmath87 , or @xmath74 , respectively.,title="fig : " ] 0.765 , or @xmath87 , as a function of @xmath87 , or @xmath74 , respectively.,title="fig : " ] ) vs the curve ( dashed ; @xmath109 ) in the @xmath108 plane at which the casimir force attains minimum for a given temperature @xmath74 .
note that for relatively small negative values of @xmath74 , the jump in the force when one crosses the phase coexistence curve and the minimal casimir force actually occur at different values of the field @xmath87 .
the last implies that by increasing the temperature one passes though the minimum of the force which then continuously increases till crossing the coexistence curve at which the force jumps to negative values much closer to @xmath110 . for @xmath111 the minimal value of the force
is practically achieved at the phase coexisting line.,scaledwidth=70.5% ] using the derived exact analytical expressions described above in the current section we determine the casimir force in the critical and in the capillary condensation regimes at temperatures below but relatively close to @xmath112 .
it should be pointed out that the solutions @xmath90 of the transcendental equation ( [ treq ] ) that correspond to certain values of the parameters @xmath74 and @xmath75 are to be obtained numerically identifying by inspection those of them that obey the conditions formulated above . the behavior of the normalized finite - size scaling function @xmath113 of the casimir force is shown in figures [ fig : dp ] and [ fig : cs ] .
the relief map of the casimir force , as a function of both @xmath74 and @xmath87 , is shown in figure [ fig : dp ] where the upper part presents the force in a larger scale , while the lower one is a blow up of the region close to the bulk critical point .
the only other model we are aware of where such a relief map as a function of both relevant scaling variables is available is that one of the three - dimensional spherical model under periodic boundary conditions @xcite .
one observes a valley in this map with its deepest point at the dashed line shown in figure [ fig : phd_vs_mcf ] .
figures [ casplusmf ] and [ fig : cf ] present cross - sections of the foregoing 3d figures for given fixed values of @xmath74 , or @xmath87 , as a function of @xmath87 , or @xmath74 , respectively
. please note that they cover region of parameters that goes beyond the one covered in figure [ fig : dp ] .
figure [ casplusmf ] shows the behavior of @xmath114 as a function of @xmath74 for @xmath115 .
note that @xmath114 is negative and for @xmath105 has a minimum at @xmath116 _ above _
@xmath117 , as in the case of the @xmath118 ising model @xcite .
the value of the minimum is @xmath119 times deeper than the corresponding value of the force at @xmath9 , which agrees with the results of @xcite .
the overall behavior of the curves is similar to that one obtained for the @xmath120 ising model @xcite via density - matrix renormalization group ( dmrg ) calculations .
let us , nevertheless , note that for bulk @xmath120 ising type systems there is no true phase transition for finite @xmath22 .
it has been shown , however , that something similar does exist a line of very weakly rounded first - order transitions ending in a pseudo - critical point @xcite .
let us note that for @xmath121 , i.e. , @xmath122 , there is a special region of values of the field parameter when @xmath123 .
for such values of @xmath87 one does not cross the capillary condensation line when lowering the temperature see figure [ fig : pd ] .
let us also note that , of course , one formally can consider temperatures well below @xmath112 but then the model will no longer deliver physically sensible information .
we recall that in the current study when taking @xmath124 we do consider temperatures below but quite close to @xmath112 .
figure [ fig : cf ] depicts the behavior of @xmath114 as a function of @xmath87 for @xmath125 .
note that the minimum of the function @xmath126 is again negative , it is attained at @xmath127 , and is @xmath128 times deeper than the corresponding value at the bulk critical point .
the markers on the curves , including the inset curve representing the blow - up in the case @xmath129 , show an excellent agreement of the numerical results obtained in @xcite ( filled markers ) , and in @xcite ( empty squares ) with the analytic results ( solid lines ) presented here .
we observe that the casimir force exhibits a discontinuous jump on crossing the coexistence line as predicted in @xcite and as also shown in @xcite .
it is easy to estimate the magnitude of this jump .
one way of arguing is through direct formal use of the kelvin equation @xcite as it is done in @xcite , which leads to the conclusion that the jump @xmath130 where @xmath131 is the bulk spontaneous order parameter , @xmath132 is the field on the capillary coexistence line and @xmath133 is the interfacial tension between the coexisting @xmath134 and @xmath135 phases of the fluid .
here we have taken into account that according to the kelvin equation for a fixed @xmath124 and large @xmath22 coexistence happen at @xmath136 $ ] @xcite .
the above expression for @xmath137 implies that the magnitude of the jump decreases in the same fashion as the interfacial tension , as @xmath0 increases at fixed @xmath22 towards @xmath112 .
since the jumps is to an exponentially small force upon crossing the capillary condensation line the above estimate for the jump of the force can be also considered as an very rough estimation of the maximal value of the force for a given @xmath0 near the line of coexistence of the two phases .
let us also note that for @xmath0 below but close to @xmath112 the above expressions for @xmath137 take a scaling form : @xmath138 and @xmath139 , as it is actually also clear from the scaling relation we have derived for that region of thermodynamic parameters .
we recall that within the mean - field approach @xmath140 and , formally , @xmath141 in the scaling relations .
we see that , as argued in @xcite , in a fluid confined by identical , strongly adsorbing walls the casimir force is strongly influenced by capillary condensation .
the dmrg results presented there ( see figure 1 in @xcite ) and in @xcite for @xmath120 ising strip subject to identical surface fields also support this .
let us note that for @xmath120 at @xmath9 the scaling function of the casimir force has a minimum about 100 times bigger than the casimir amplitude @xcite .
we conclude that our results given in figures [ fig : dp ] , [ casplusmf ] and [ fig : cf ] are in full conformity with the statement made in @xcite that in a fluid confined by identical , strongly adsorbing walls the casimir force is much stronger for states which lie slightly off bulk coexistence , with @xmath142 .
we have derived exact analytical results for the thermodynamic casimir force , see eq .
( [ casimir_final ] ) , in a widely used model in the theory of phase transitions . in this model ,
the value @xmath90 of the order parameter in the middle of the system is a solution of eq .
( [ treq ] ) obeying the condition ( [ eq : condition ] ) .
if there is more than one solution @xmath99 satisfying the above requirements , we take the one that leads to an order parameter profile corresponding to the minimum of the energy functional ( [ afunctional ] ) , as explained in details in @xcite .
the obtained results allow us to plot the relief map of the force , see figure [ fig : dp ] , as a function of the both relevant scaling variables the temperature and field .
in addition , figures [ casplusmf ] and [ fig : cf ] present cross - sections of the behavior of the force for given fixed values of @xmath74 , or @xmath87 , as a function of @xmath87 , or @xmath74 , respectively .
finally , figure [ fig : phd_vs_mcf ] presents the loci of the minima of the force in the @xmath108 plane .
the inspection of the results convincingly demonstrates that for @xmath142 the casimir force in the capillary condensation regime is much more attractive than that found at the critical point .
the comparison there of the numerical evaluation of our analytical expressions with the available numerical results reported in @xcite shows an excellent agreement between each other .
furthermore , the analysis of the available dmrg numerical results @xcite for the behavior of the casimir force for the 2d ising model , as presented in sect .
[ sec : num_results ] , led us to conclusion that our mean - field results , which formally correspond to systems with spacial dimension @xmath143 , are quite similar in the capillary condensation regime to those of the two - dimensional ising model .
thus , the results for the capillary condensation regime are quite robust with respect to the influence of the dimensionality .
of course , the situation is different in the parametric space close to the bulk critical point , as well as very close to the critical point of the finite system .
we recall that this later critical point shall show the singularities of the corresponding @xmath144-dimensional system . within the mean - field approach , as it is well known ,
the role of the fluctuations is not properly taken into account . if one wants to go beyond the mean - field approximation for , say , @xmath145 dimensional systems one shall either relay on numerical methods like in @xcite , or use methods based on renormalization group approach
. however , let us recall that the mean - field results serve as a starting point there for renormalization group calculations @xcite .
thus , our results shall be helpful for such future analytical studies on the thermodynamic casimir force .
finally , let us also remind that in physical chemistry and , more precisely , in colloid sciences fluid mediated interactions between two surfaces or large particles are usually referred to as solvation forces or disjoining pressure @xcite . thus , our results can be also considered as pertaining to a particular case of such forces when the fluid is near its critical point .
h b g 1999 some remarks on the history of the so called casimir effect _ the casimir effect 50 years later _ proceedings of the fourth workshop on quantum field theory under the influence of external conditions , leipzig , 1998 ed bordag m ( world scientific ) pp 39 french r h , parsegian v a
, podgornik r , rajter r f , jagota a , luo j , asthagiri d , chaudhury m k , chiang y m , granick s , kalinin s , kardar m , kjellander r , langreth d c , lewis j , lustig s , wesolowski d , wettlaufer j s , ching w y , finnis m , houlihan f , von lilienfeld o a , van oss c j and zemb t 2010 _ rev .
phys . _ * 82*(2 ) 18871944 | when massless excitations are limited or modified by the presence of material bodies one observes a force acting between them generally called casimir force .
such excitations are present in any fluid system close to its true bulk critical point .
we derive exact analytical results for both the temperature and external ordering field behavior of the thermodynamic casimir force within the mean - field ginzburg - landau ising type model of a simple fluid or binary liquid mixture .
we investigate the case when under a film geometry the boundaries of the system exhibit strong adsorption onto one of the phases ( components ) of the system .
we present analytical and numerical results for the ( temperature - field ) relief map of the force in both the critical region of the film close to its finite - size or bulk critical points as well as in the capillary condensation regime below but close to the finite - size critical point .
_ keywords _ : rigorous results in statistical mechanics , classical phase transitions ( theory ) , finite - size scaling |
the concept of coherent control precise measurement or determination of a process through control of the phase of an applied oscillating field has been applied to many different systems , including quantum dynamics@xcite , trapped atomic ions@xcite , chemical reactions@xcite , cooper pairs@xcite , quantum dots@xcite and thz generation@xcite to name but a few .
a plasma wave is a coherent and deterministically evolving structure that can be generated by the interaction of laser light with plasma .
it is therefore natural to assume that coherent control techniques may also be applied to plasma waves .
plasma waves produced by high power lasers have been studied intensively for their numerous applications , such as the production of ultrashort pulses by plasma wave compression@xcite , generation of extremely high power pulses by raman amplification@xcite , for inertial confinement fusion ignition schemes@xcite , as well as for fundamental scientific investigations . in particular ,
laser wakefield acceleration of ultra - relativistic electron beams@xcite , has been a successful method for accelerating electrons to relativistic energies over a very short distance . in laser wakefield acceleration , an electron bunch ` surfs ' on the electron plasma wave generated by an intense laser and gains a large amount of energy .
the accelerating electric field strength that the plasma wave can support can be many orders of magnitude higher than that of a conventional accelerator , which makes laser wakefield acceleration an exciting prospect as an advanced accelerator concept .
however , although highly competitive in terms of accelerating gradient , beams from laser wakefield accelerator experiments are currently inferior to conventional accelerators in terms of other important characteristics , such as energy spread and stability .
in addition , due to constraints in laser wakefield technology , experimental demonstrations have predominantly been performed in single shot operation , far below the khz - mhz repetition rates of conventional accelerators . in recent years
, deformable mirror adaptive optical systems have been successfully implemented in high intensity laser experiments to increase the peak laser intensity by improving the beam focusability , especially in systems using high numerical aperture optics .
the shape of the deformable mirror is generally determined in a closed loop where either a direct measurement of the wavefront is performed@xcite or some nonlinear optical signal@xcite is used as feedback in an iterative algorithm .
the objective of adaptive optics has largely been optimization of the laser focal shape to a near diffraction - limited spot , thus producing the highest possible intensity .
adaptive optics can also be useful for certain focal profile shaping@xcite , optimization of a laser machining process@xcite or harmonic generation@xcite . in the following ,
we demonstrate that orders of magnitude improvement to electron beam properties from a laser wakefield accelerator operating at khz repetition rate can be made , through the use of a genetic algorithm coupled to a deformable mirror adaptive optical system to coherently control the plasma wave formation . the electron image from a scintillator screen
was processed and used in the fitness function as feedback for the genetic algorithm . using this method
, we were able to improve the beam properties significantly .
this result was not simply due to an improvement in focal quality since a laser pulse with the ` best ' ( highest intensity / lowest @xmath0 ) focus in vacuum produced a greatly inferior electron beam compared with a laser pulse optimized using the electron beam properties themselves .
it was found that the focal spot optimized for electron beam production had pronounced intensity ` wings ' .
modifications to the phase front of the tightly focusing laser alter the light propagation , which experiences strong optical nonlinearities in the plasma , and therefore affect the plasma wave dynamics in a complex but deterministic manner .
the experiment was performed using the relativistic lambda - cubed ( @xmath1 ) laser system ( see methods ) .
the output laser beam was reflected from a deformable mirror and focused onto a free - flowing argon gas plume to produce an electron beam by laser wakefield acceleration ( see methods ) at 500 hz .
electrons were measured using a scintillating screen imaged onto a lens - coupled ccd camera .
the experimental setup is shown schematically in fig .
[ setup ] .
we first implemented a genetic algorithm for laser focus optimization using the second - harmonic signal generated from a beta barium borate ( @xmath2-bbo ) crystal ( setup a in fig .
[ setup ] ) .
the laser spot was optimized such that highest peak intensity is achieved when the second harmonic generation is strongest .
subsequently , we modified the fitness function to use a figure of merit ( fom , refer to equation [ fom2 ] in methods ) from the electron scintillation data , calculating the inverse distance weighting ( with power parameter n ) to a single point @xmath3 for all pixel intensities within an electron image .
the pixel of the optimization point @xmath3 was _ dynamically _ adjusted during the genetic algorithm to concentrate all electron signal to the peak location of the charge distribution during each generation .
the genetic algorithm was initialized using a ` flat ' mirror shape with 30 v for all actuators to allow immediate deformation in both directions . for comparison , electron beams produced by the ` best ' laser focus ( by optimizing the intensity ) and the initial mirror shape at 30 v
are shown in fig .
[ ebeam]a and b respectively .
the optimized electron beam profiles are shown in fig .
[ ebeam]e - j for various weighting parameters , @xmath4 .
the genetic algorithm converged to the best electron beam using @xmath5 in terms of beam divergence and peak charge density .
the peak charge density was increased by a factor of 20 compared to the initial electron beam profile before optimization ( see fig . [ ebeam]d ) .
the optimized electron profile is highly stable and collimated , with a full - width - at - half - maximum ( fwhm ) divergence of @xmath6 mrad and @xmath7 mrad .
the shot - to - shot pointing ( defined by the centroid position ) fluctuation of the electron beam is less than 1 mrad ( root mean square , r.m.s . ) .
the integrated charge was increased by more than two - fold from the electron beams generated by a laser focus of highest intensity .
the high repetition rate and real - time diagnostics permit implementation of the algorithm within a practical time frame using a standard personal computer .
typical optimization takes only a few minutes ( @xmath840 iterations ) to reach convergence ( see fig . [
ebeam]c ) .
the second harmonic optimization@xcite generates a near - diffraction - limited focal spot as shown in fig .
[ spot]a for the far - field laser intensity profile in vacuum . in figs .
[ spot]a and b , we compare the transverse intensity distribution within the focal region over the length scale of the gas jet .
the laser profile ( fig .
[ spot]b ) that produces the best electron beam exhibits several low intensity side lobes around the central peak , and has a peak intensity about half that of the optimized focus .
the complex laser profiles appear to have a very dramatic effect on the structure of the plasma waves produced and consequently the electron beam profile .
[ spot]c shows the relative wavefront change recorded by a shack - hartmann wavefront sensor .
calculation using the reconstructed wavefront gives a strehl ratio of about 0.5 , which is in agreement with the far - field intensity measurement .
this small wavefront modification of the driver pulse can lead to a significant improvement in the electron beam properties through the relativistic nonlinear optics of the plasma .
the relative position between the focal plane and the center of the gas flow was controlled by moving the nozzle .
scanning the gas nozzle both before and after genetic algorithm confirms the optimal focal position does not change , excluding the possibility that the improvement may be due to optimizing the focal position .
furthermore , we extended the genetic algorithm optimization to _ control the electron energy distribution_. through control of the light propagation , the plasma wave amplitude will be affected and therefore also the strength of the accelerating gradient .
hence , we can expect to be able to modify the energy spectrum .
a high resolution energy analyzer using a dipole magnet pair was used to obtain the electron energy spectrum as the electrons were dispersed in the horizontal plane in the magnetic field .
a 150 @xmath9 m pinhole was placed 2.2 cm from the electron source to improve the energy resolution of the spectrometer .
the schematic setup is shown in fig .
[ espec]a .
the energy resolution limited by the entrance pinhole and transverse emittance of the beam is estimated to be 2 kev for the energy range of measurement .
three rectangular masks are set in the low- , mid- and high - energy region on the dispersed data , namely masks i , ii , and iii in fig .
[ espec]b .
we employed a fitness function ( see methods ) to preferentially maximize the total counts inside the mask .
raw spectra from the genetic algorithm optimization are displayed in fig .
[ espec]b , showing that the brightest part has shifted congruently .
the resulting spectra have mean energies of 89 kev , 95 kev and 98 kev respectively for masks i , ii , and iii , noting that they do not fall on the visual centroid of the image because the scintillator sensitivity is not included in the presentation of the raw data , however it was taken into account for computing the mean energies .
our results show that manipulation of electron energy distribution using the deformable mirror is somewhat restricted .
the final result after optimization does not reach the objective mask completely despite that the mean energies can be varied by up to 10% .
this result is somewhat unsurprising as the scope for controlling the electron spectrum is mostly limited by the physics of the interaction - while changing the transverse intensity profile can make big differences to the shape of the plasma electric field structure , changing the maximum field amplitude of the wakefield ( and therefore peak energy of the accelerated electrons ) will be limited .
although the details of the initial conditions required for optimal beams are difficult to determine and are therefore found using the genetic algorithm , we can at least demonstrate how modifications to the phase front of the laser pulse can improve the beam properties with an example . to illustrate the underlying physics of the plasma wave dynamics determined by the conditions of the driving laser pulse , we performed two dimensional ( 2d ) particle - in - cell ( pic ) simulations using the osiris framework@xcite . parameters similar to the experiment conditions were used , with a gaussian plasma density profile to enable trapping of electrons in the density down ramp ( see ref . and methods for details on the simulation ) .
it was previously shown in ref . that the focusing fields of laser plasma accelerators can be controlled by tailoring the transverse intensity profile of the laser pulse using higher - order modes , where generalization to 3d was also discussed . here , we simulated a laser pulse with a fundamental gaussian mode ( tem@xmath10 ) or a coherent superposition of a fundamental ( tem@xmath10 ) and a second - order hermite - gaussian ( tem@xmath11 ) mode ( fig .
[ sim]a ) .
although the plasma wave has a larger amplitude when it is driven by a single mode laser pulse , the wake phase front evolves a backward curvature when electrons are trapped and accelerated ( see top panel in supplementary movie 1 and fig .
[ sim]b ) .
contrastingly , the evolution of the wakefield driven by the laser pulse with additional mode forms a flatter plasma phase front _ at the point of trapping _ ( fig .
[ sim]d ) . in fig .
[ sim]c and e , the momentum distribution of the forward accelerated electrons has a larger transverse spread for the single mode laser pulse compared to the one with the addition of higher order modes .
this is a consequence of the different trapping conditions and accelerating fields from the coherent plasma wakefield structure , which is governed by the structure of the driving laser pulse . in a comparative test
to show this effect is not simply due to a lower intensity , we repeated the simulation using a single fundamental gaussian mode laser pulse with a larger focal spot with the same peak intensity as that with the superimposed modes .
the wakefield evolution shows very similar response as fig .
[ sim]b and does not develop a flatter phase front as seen in fig .
the subsequently accelerated electrons have very similar divergence to that in fig .
[ sim]c , eliminating the possibility that the improvement comes from either a high intensity effect or a simple change in @xmath12-number .
_ note that we are not saying that this mixed - hermite - gaussian mode is the optimal pulse in the experiment ; this is simply an illustration of how small changes in pulse shape can have significant effects on electron beam properties .
when a particular wavefront of laser light interacts with plasma , it can affect the plasma wave structures and trapping conditions of the electrons in a complex way .
for example , raman forward scattering , envelope self - modulation , relativistic self - focusing , and relativistic self - phase modulation@xcite and many other nonlinear interactions modify both the pulse envelope and phase as the pulse propagates , in a way that can not be easily predicted and that subsequently dictates the formation of plasma waves .
moreover , under realistic experimental conditions , ionization dynamics before the laser pulse reaches the vacuum focus can also modify the phase of the driving pulse . ideally , the light interacts in such a way as to generate large amplitude plasma waves with electric field structures that accelerate electrons with small divergence , high charge etc . because of the complicated interaction , it is difficult to determine a laser phase profile that will lead to such a plasma structure .
however , such unforeseeable conditions were successfully revealed by _ using the evolutionary algorithm method _
, with the result that the electron charge can be increased and emitted in a very well collimated beam .
here we have implemented coherent control of a nonlinear plasma wave and demonstrated an order of magnitude improvement in the electron beam parameters .
the laser beam optimized to generate the best electron beam was not the one with the ` best ' focal spot .
control and shaping of the electron energy distribution was observed to be less effective , but was still possible .
the capability for wavefront control was also limited by the number of actuators and maximum deformation of the deformable mirror used in our experiments .
in addition , this work was performed using adaptive optics , but it is clear that coherent control of plasma waves should be possible in a variety of configurations , for example by using an acousto - optic modulator to control the temporal phase of the driving pulse .
recently developed techniques@xcite for single - shot diagnosis of plasma wave structures may provide an avenue for direct control of the plasma evolution .
the concept of coherent control for plasmas opens new possibilities for future laser - based accelerators .
although still at the stage of fundamental research , laser wakefield accelerators are showing significant promise . in principle , such improvements could be integrated into next generation high - power laser projects , such as ican@xcite , based on coherent combination of many independent fibers , taking advantage of both their high repetition rate and controllability .
the stability and response of the wakefield to laser conditions , such as phase front errors , is not well understood , but is crucial for the success of laser wakefield acceleration as a source of relativistic electrons and secondary radiation .
for example , the presence of an asymmetric laser pulse was shown to affect the betatron oscillations and properties of x - rays produced in laser wakefield accelerators@xcite .
implementing the methods of this study should enable a significantly improved understanding and control of the wakefield acceleration process with regard to stability , dark current reduction and beam emittance .
the relativistic lambda cubed laser ( @xmath1 ) produces 30 fs pulses of 800 nm light at a repetition rate of 500 hz with an ase ( amplified - spontaneous - emission ) intensity contrast of @xmath13 around 1 ns before the main pulse .
the system is seeded by a femto - laser ti : sapphire oscillator , which generates 12 fs pulses and has a companion carrier envelope phase locking system .
an rf addressable acousto - optic filter called a dazzler controls the spectral amplitude and phase of these pulses . selected pulses from the dazzler train
are stretched to 220 ps in a low - aberration stretcher and amplified to 7 mj in a cryogenically cooled large - mode regenerative amplifier ( regen ) .
the energy dumped from the regen cavity is ` cleaned ' in a pockels cell and used to seed a 3-pass amplifier as an upgrade from the laser system described in ref . , which delivers up to 28 mj pulses before compression .
following 71% efficient compression , 20 mj pulses are trimmed to 18 mj at the perimeter of a 47 mm - diameter , 37-actuator deformable mirror . throughout the system ,
pump light is provided by a variety of internally doubled nd - doped yag , ylf and vanadate lasers . the output beam with its controllable wavefront
is then delivered to one of five experimental areas for the production of x - rays , electron beams , ion beams , thz radiation , high - order harmonics , or warm - dense matter .
the focused laser pulse drives plasma waves by interacting with an argon gas jet flowing continuously from a 100 @xmath9 m inner diameter fused silica capillary .
typically the laser axis is 300 @xmath9 m above the orifice of the tubing .
the laser pulses were focused by an @xmath14 off - axis parabolic mirror to a spot size of 2.5 @xmath9 m fwhm with a maximum of 10 mj energy on target .
the plasma electron density is measured to be in the range ( 0.5 - 2)@xmath15 @xmath16 using transverse interferometry .
electrons are accelerated in the density down ramp with a final energy in the 100 kev range@xcite and detected by a high resolution scintillating screen ( j6677 fos by hamamatsu ) , which is placed about 35 cm downstream from the source and imaged with a lens coupled 12-bit ccd camera for a 4@xmath174 cm effective area .
the scintillator sensitivity was calibrated using an electron microscope for the energy range in the spectrum measurement .
electron beam charge was estimated using the calibrated scintillator response , manufacturer - provided information for the ccd camera ( gain , quantum efficiency etc . ) and the measured effective numerical aperture of the imaging system .
the amplified laser beam was attenuated by using a half - wave plate and the polarization dependent properties of the compressor grating of the laser system .
25 mm neutral density filters ( thorlabs , inc . )
were inserted in the exit beam after the compressor and before a telescope beam expander .
the laser focus was imaged by a 60@xmath17 microscope objective lens ( newport corporation , m-60x ) onto a 8-bit ccd camera for focal characterization ( cf setup a in fig .
[ setup ] ) . a shack - hartmann wavefront sensor ( flexible optical bv )
was used to determine the relative wavefront change between different deformable mirror configurations .
the sensor , which consists of a @xmath18 microlens array having a focal length of 3.5 mm and 150 @xmath9 m pitch , was directly placed in the path of the converging beam after the focusing parabolic mirror .
reconstruction of the wavefront was performed using the frontsurfer analysis software ( flexible optical bv ) , typically with @xmath19 measured local wavefront slopes and rms error on the order of 0.05@xmath20 .
rotating the neutral density filters and the half - wave plate did not change the focal spot or the wavefront measurement significantly , insuring the wavefront distortion introduced by attenuation was negligible .
the deformable mirror ( aoa xinetics ) has a 47-mm clear aperture of a continuous face sheet with 37 piezoelectric actuators arranged on a square grid spaced 7 mm apart .
the maximum stroke used in this experiment is about 2 @xmath9 m .
the mirror shape is controlled by a genetic algorithm , which is a method mimicking the process of natural selection and routinely used to generate optimal solutions in complex systems with a large number of variables .
the _ genetic representation _ in our experiments comprises a set of 37 independent voltage values for the deformable mirror actuator array .
a _ fitness function _ is designed to produce a single _ figure of merit _ ( fom ) to evaluate how close the solution is to the goal . in the electron beam profile optimization experiment , fom is computed as follows : @xmath21 where @xmath22 is the pixel intensity for every pixel @xmath23 in the whole image and @xmath3 is a coordinate point in the image used as an optimization target . the power factor
@xmath24 gives higher weighting to those pixels closer to the target ( inverse distance weighting ) . in the experiment to control the energy spectrum , fom
is calculated using the following formula given a pre - defined image mask , @xmath25 the mean intensity is the sum of the pixel counts divided by the number of pixels for a defined region . a rectangular mask was used in the experiment as specified by the region enclosed by the red dashed lines in fig .
[ espec]b .
the 2d pic simulations were performed in a stationary box of the dimensions @xmath26 @xmath9 m with @xmath27 cells and 4@xmath174 particles - per - cell .
a gaussian plasma density profile was used in the propagation dimension ( @xmath28 ) , peaked at @xmath29 m with a full - width - at - half - maximum of 120 @xmath9 m and a maximum electron density of 0.005@xmath30 , where @xmath30 is the plasma critical density .
the laser pulse was initialized at the left edge of the simulation window and focused at 215 @xmath9 m in the density down ramp . in 2d geometry ,
the transverse intensity profile of the laser pulse for fundamental gaussian mode ( tem@xmath10 ) has the form @xmath31 , and the second - order hermite - gaussian mode @xmath32 , where @xmath33 is the normalized vector potential and @xmath34 is the beam waist parameter .
the two modes are coherently superimposed in the same plane of polarization . here
we used even - order hermite - gaussian mode ( tem@xmath11 ) for its symmetric property .
a phase difference of @xmath35 was applied at the beam waist between the two modes to simulate variations in the optical phase front condition .
the beam waist was positioned to account for focal shift as a result of coherent superposition of two modes such that the location of the maximum on - axis laser intensity was the same ( at @xmath36 m ) for all simulation runs .
the simulation parameters are @xmath37 and @xmath38 @xmath9 m for the gaussian mode alone , or @xmath39 , @xmath40 and @xmath41 @xmath9 m for the superimposed mode .
this work was supported by darpa ( contract no .
n66001 - 11 - 1 - 4208 ) , the nsf ( grant no .
0935197 ) , nsf career ( grant no 1054164 ) , the afosr young investigator program ( grant no . fa9550 - 12 - 1 - 0310 ) and mcubed at the university of michigan .
the authors acknowledge the osiris consortium , consisting of university of california , los angeles ( ucla ) and instituto superior tcnico ( ist ) ( lisbon , portugal ) , for the use of the osiris 2.0 framework . this research was supported in part through computational resources and services provided by advanced research computing at the university of michigan , ann arbor . z.h , b.h . , j.a.n
designed and carried out the experiment .
z.h . performed all data analysis and the simulations .
b.h . developed the labview program .
k.m.k and a.g.r.t . directed and guided the project .
z.h . and a.g.r.t .
wrote the paper .
all authors discussed the results and contributed to the manuscript . .
( c ) relative wavefront change reconstructed from direct measurement using a shack - hartmann wavefront sensor .
root - mean - square phase deviation in the aperture is 0.14 wave .
the wavefront was reconstructed over a slighly smaller aperture ( 2.26 mm diameter ) than the full beam diameter on the sensor ( @xmath422.7 mm @xmath43 width ) to reduce errors in the peripheral area .
( scale bar , 1 mm).,width=160 ] @xmath9 m ) for a single gaussian mode ( solid curves ) and a superimposed mode ( dashed curves ) obtained in the 2d simulations . plotted are the laser pulses at vacuum focus .
( b)(d ) are snapshots showing the different plasma wave structures around the time ( @xmath44 ps ) and location of trapping , driven by a laser pulse of single mode ( b ) and superimposed mode ( d ) .
the laser pulse propagates to the right .
the phase space @xmath45 distribution of the accelerated electrons shown for ( c ) single mode and ( e ) superimposed mode before the electrons exit the simulation box ( @xmath46 ps).,width=160 ] | coherent control of a system involves steering an interaction to a final coherent state by controlling the phase of an applied field .
plasmas support coherent wave structures that can be generated by intense laser fields . here
, we demonstrate the coherent control of plasma dynamics in a laser wakefield electron acceleration experiment .
a genetic algorithm is implemented using a deformable mirror with the electron beam signal as feedback , which allows a heuristic search for the optimal wavefront under laser - plasma conditions that is not known _ a priori_. we are able to improve both the electron beam charge and angular distribution by an order of magnitude .
these improvements do not simply correlate with having the ` best ' focal spot , since the highest quality vacuum focal spot produces a greatly inferior electron beam , but instead correspond to the particular laser phase front that steers the plasma wave to a final state with optimal accelerating fields .
center for ultrafast optical science , university of michigan , ann arbor , mi 48109 - 2099 usa polytech paris - sud - universit paris - sud , 91405 orsay , france |
predicting the integrated spectral energy distributions of stellar populations is important in the solution of many problems in astronomy , from determining the ages of globular clusters to modeling counts of faint galaxies at high redshift . beginning with the early work of @xcite , successive generations of modelers
have combined the best available data on stellar structure and evolution to predict the appearance of the combined light of generations of stars .
although the subject of population synthesis has a long history , it is an active area of research : synthesis techniques and many of the input data ( isochrones , opacities , spectral libraries ) continue to be improved
. there is good evidence that globular clusters ( gcs ) are internally homogeneous in age and metallicity @xcite .
gcs are the best observational analogs of modelers ` simple stellar populations ' , i.e. populations of stars formed over a short time out of gas with homogeneous chemical composition .
broadband colors are among the simplest predictions of population synthesis models , so comparing the models predicted colors to cluster colors is the natural zeroth - order test of compatibility between the models and reality @xcite . in this paper
we compare the broad - band @xmath0 colors predicted by three modern population synthesis models with the colors of galactic and m31 gcs .
our observational database is the first one to include extensive coverage of the @xmath5 bandpasses , and the first with spectroscopic metallicities for all clusters .
we use the cluster metallicities to bin the clusters for comparison to the appropriate models . in this way
we determine the cluster - to - model offset separately for each color and avoid the ambiguity in comparing model and cluster colors in two - color diagrams .
for m31 clusters , the observational data are from the @xcite catalog . for the galactic clusters we obtained optical colors , metallicity , and reddening from the june 1999 version of the @xcite catalog , and ir colors from @xcite , as reported in @xcite ( but with the reddening correction applied by @xcite removed and the reddening values in @xcite used instead ) .
we dereddened the clusters colors using the values of @xmath6 given in the catalogs and the @xcite extinction curve for @xmath7 . for m31
, we excluded clusters where the error in the spectroscopic metallicity was @xmath8}>0.5 $ ] , and clusters suspected of being young on the basis of strong balmer absorption or blue colors ( see * ? ? ?
* ) . for both galaxies , we excluded clusters with @xmath9 ; there are 103 m31 and 85 galactic clusters in the final sample .
photometric data is not available in all bandpasses for all clusters : only about two - thirds have measured @xmath10 and @xmath11 , and less than half have @xmath12 .
we compare the cluster colors to those for simple stellar populations of ages 8 , 12 , and 16 gyr from three sets of models : those of worthey}<-1.0 $ ] .
] , bruzual and charlot ( hereafter bc ) ( both the worthey and bc models are reported in * ? ? ?
* ) , and @xcite ( hereafter kff ) .
although model colors are tabulated in smaller age increments ( typically 1 gyr ) , initially it is more reasonable to use the models as a rough guide to relative ages rather than attempting to derive precise cluster ages from them .
the worthey models are computed at [ fe / h ] values of @xmath13,@xmath14,@xmath15,@xmath16 , and @xmath17 dex , and the bc and kff models are computed at [ fe / h ] values of @xmath18 ( kff models only ) , @xmath19 , @xmath20 , and @xmath21 dex .
we compared clusters to both the salpeter imf ( worthey s ` vanilla ' models ) and @xcite ( * ? ? ? * in the worthey models ) imf version of the models .
@xcite finds that some of his model colors have defects ( e.g. is too red by 0.04 - 0.06@xmath22 due to problems in the theoretical stellar atmospheres and the color - temperature calibration ) , but the sizes of these defects are not well - determined so we do not correct for them .
figure [ twocolor ] shows data and models in two frequently - used two - color diagrams .
since the models are computed at discrete values of [ fe / h ] , we use the spectroscopic metallicities of the clusters to compare only clusters with comparable metallicities ( @xmath23 dex ) to each model .
the galactic cluster metallicities given in @xcite are on the @xcite ( zw ) metallicity scale , and the m31 cluster metallicities are also tied to this scale through the calibration of @xcite .
recent work @xcite suggests that the zw scale may be non - linear at both high and low metallicities .
we retain the zw scale in this paper because we found that using the @xcite scale to assign clusters to model comparison bins made little difference in our results .
we caution , however , that the effect of changing the metallicity scale is unknown for the @xmath4}=-0.25 $ ] model bin .
the transformation from the zw to cg scales is only defined for @xmath4_{zw}}<-0.5 $ ] , the lower limit of this metallicity bin .
we calculated the mean offsets between model and cluster colors ( referenced to @xmath24 ) for each metallicity bin ; figures [ salp - zwscale][scal - zwscale ] show some representative comparisons .
we plot @xmath25 for all bandpasses @xmath26 to make clear the differences in spectral energy distributions between models and data ; we remind the reader that the offsets for bandpasses redward of @xmath24 thus have the opposite sign from the usual colors .
one general characteristic of the models visible in the figures is that younger - aged models predict bluer colors .
the exception is the kff scalo model for @xmath4}=-1.63 $ ] , which predicts only very small color differences ( @xmath27 ) between ages of 12 and 16 gyr .
the effect of the imf on the colors appears to depend on both metallicity and age . for the worthey @xmath28}=-1.50 $ ] models , miller - scalo imf colors are redder than salpeter model colors at all ages , but for the @xmath28}=-0.50 $ ] models , the miller - scalo imf colors are bluer for 8 and 12 gyr and almost identical for 16 gyr .
bc predict almost no color difference between the salpeter and scalo imf models of the same age and metallicity .
a striking feature in figures [ salp - zwscale ] and [ scal - zwscale ] is the range of discrepancies between models and data .
for example , the largest difference between the worthey model with parameters ( salpeter imf , @xmath28}=-1.50 $ ] , age 16 gyr ) and the mean colors of clusters with @xmath29}\leq-1.25 $ ] is @xmath30 in . the same models with
@xmath28}=-0.50 $ ] are well offset from the data at all colors except ; the largest offset is @xmath31 in . to determine the best - fitting models
, we quantify the overall goodness - of - fit for each model / cluster metallicity bin pair as : @xmath32 the color differences @xmath33 are weighted by @xmath34 , where @xmath35 are the standard errors in the mean colors of objects in the bin .
table [ bestfit ] gives the @xmath36 and @xmath37 values for the best fitting models in each metallicity bin .
table [ bestfit ] shows that the best - fitting models fit the data quite well , with typical color offsets of @xmath38 .
the two bandpasses with the most significant offsets are @xmath2 and @xmath3 : the models are too blue in and too red in .
neither offset shows a clear trend with metallicity .
the offsets are likely not due to systematics in the photometric system or in the extinction curve . while problems with the photometric systems might be expected in the @xmath10 and @xmath11 bands ( due to conversion between the johnson and cousins @xmath39 systems ) , both data and models use the well - defined johnson system .
problems in the reddening curve also seem unlikely for the same reasons .
we suspect that the offsets are more likely due to systematic errors in the models .
the offset in particular is likely due to the flux libraries used .
both @xcite and @xcite found their model colors to be @xmath40 too red compared to empirical solar - metallicity spectra , even after correcting to the empirical color - temperature scale .
this suggests a possible problem with the stellar atmosphere models of @xcite , upon which both libraries are based .
the cause of the offset in is not as clear .
this offset is actually worse than it appears : since we compute the colors for the bc and kff models as @xmath41 , the red colors _ compensate _ for some of the defect , which is actually larger than the defect in .
whose models give directly finds that his model is too blue compared to solar neighborhood stars and elliptical galaxies .
worthey cites problems with the @xmath2 fluxes from the stellar libraries as a possible cause : modeling the many blended atomic and molecular lines blueward of @xmath3 is difficult , and many of the necessary opacities are not well determined .
this can not be the only cause of model problems in @xmath2 , since the bc and kff models , which use the same stellar library , predict different colors .
the treatment of the horizontal branch in the models is another possible source of problems in the colors because the hb emits most of the blue light .
however , systematic problems with the model hb color ( which depends on metallicity ) , would presumably produce a offset dependent on metallicity , which we do not observe .
observational error is another possible contributor to the offset , as many of the colors of the m31 clusters are poorly determined ( see table 3 of * ? ? ?
understanding the rest - frame @xmath2 flux of stellar populations becomes increasingly important when studying high - redshift galaxies and global star formation history , and further investigation of the models in this bandpass is clearly warranted .
a secondary result in table [ bestfit ] is that age determines which model best fits the data .
higher - metallicity cluster colors are best fit by 8 gyr models , regardless of imf . lower - metallicity cluster colors
( @xmath4_{bin}}\leq-1.00 $ ] ) are best fit by 12 or 16 gyr models .
the best - fit age depends on the imf for several of the models , but not in any systematic fashion .
this result is consistent with the determinations of relative ages for galactic clusters by @xcite .
these authors determined relative ages of 35 galactic globular clusters from a homogeneous set of @xmath24 , color - magnitude diagrams .
they compared theoretical isochrones with the observational cmds to determine ages using two independent methods .
they found that the clusters with @xmath4_{cg}}>-0.9 $ ] were @xmath42% younger than clusters with @xmath4_{cg}}<-1.2 $ ] , with the intermediate - metallicity clusters showing a @xmath43% age dispersion .
these results are model - dependent , as are ours , but the results similarity implies that either there is a real difference between metal - rich and metal - poor clusters or there is a systematic problem in the models in one of the metallicity regimes .
what possible systematic errors in our input data or comparison procedure could produce the result that the metal - rich clusters are younger ?
we redid the comparison procedure considering the clusters of each galaxy separately , and still found younger ages for the most metal - rich clusters . although m31 has a greater proportion of the metal - rich clusters , younger ages are found for both m31 and galactic metal - rich clusters .
@xcite suggest that the spectroscopic metallicities of the most metal - rich m31 clusters measured by @xcite are systematically too high .
if this is true , the clusters would appear too blue compared to old , higher - metallicity models and the best - fit model would be younger .
we compared the metal - rich m31 clusters to the worthey @xmath4}=-1.0 $ ] models , and the best - fitting model had age 16 gyr .
however , the goodness - of - fit was better for the young , metal - rich models than for the older , more metal - poor model , so we conclude that younger ages are still favored for these clusters . overestimating the reddening of the metal - rich clusters would make the derived intrinsic colors too blue and yield younger ages .
this seems unlikely , given that the color - metallicity relations for galactic and m31 clusters match well throughout their metallicity range ( see * ? ? ?
* ) , and the methods of reddening determination for m31 and galactic clusters are different
. if the detection of younger ages for metal - rich globular clusters is real , it has implications for galaxy formation .
a range of gc ages implies that gc formation took place over an extended period of time .
conditions for gc formation were not particular to the early universe , an assertion supported by observations of ` proto - globular ' clusters in present - day merging galaxies ( e.g. * ? ? ? * ) .
more precise knowledge of the distribution of cluster ages in each galaxy would be extremely useful in understanding cluster system formation .
if the age distribution is continuous , the relation between age and metallicity might hold clues as to what factors controlled the cluster formation rate . if the age distribution is bimodal with most clusters old and coeval and the remainder younger and coeval
then some event must have triggered the second episode of gc formation .
perhaps the younger clusters were stripped from or accreted along with satellite galaxies of m31 and the galaxy .
comparison of three sets of population synthesis models with integrated colors of m31 and galactic globular clusters shows that the models reproduce the redder average cluster colors to within the observational uncertainties .
the poorer agreement in and is likely due to systematic errors in the models .
younger - age models are required to best match the colors of the metal - rich clusters , consistent with the findings of @xcite that the most metal - rich galactic clusters are younger than the bulk of the globular cluster population .
a range of ages for globular clusters implies that conditions for cluster formation were not restricted to the early universe .
the cluster age distribution has important implications for galaxy and globular cluster system formation , and attempts to determine it more precisely are needed . vs. @xmath44 and @xmath45 for m31 globular clusters ( triangles ) and galactic gcs ( squares ) .
lines are population synthesis models of ages 8 gyr ( bluer colors ) and 16 gyr ( redder colors ) : worthey ( solid ) , bc ( dashed ) , kff ( dotted).[twocolor ] ] ( data@xmath46models ) for salpeter imf .
( @xmath47 is plotted to emphasize that the models are normalized to the data at @xmath24 . )
solid lines : 16 gyr models , dotted lines : 12 gyr models , dashed lines : 8 gyr models .
the @xmath14 and @xmath16 bins in [ fe / h ] are worthey models ; the @xmath19 and @xmath20 [ fe / h ] bins are bc models .
error bars ( plotted only on the 8 gyr models for clarity ) are the standard errors of the mean cluster colors and do not include observational uncertainties .
[ salp - zwscale ] ] or @xmath48 ( data@xmath46models ) for scalo or miller - scalo imf .
symbols same as figure [ salp - zwscale ] .
the @xmath14 and @xmath16 bins in [ fe / h ] are worthey models ; the @xmath19 and @xmath20 [ fe / h ] bins are kff models ( which have no @xmath49 since they do not predict these colors).[scal - zwscale ] ] llllrrrrrrr -0.25&w8sc & 0.023&15&@xmath50&@xmath51&@xmath52&@xmath53&@xmath54&@xmath55&@xmath56 + -0.32&k8sc & 0.022&17&@xmath57&@xmath58&@xmath59&@xmath60&&&@xmath61 + -0.50&w8sc & 0.013&30&@xmath62&@xmath63&@xmath64&@xmath65&@xmath66&@xmath67&@xmath68 + -0.63&k8sp & 0.012&35&@xmath69&@xmath70&@xmath71 & @xmath72&&&@xmath73 + -1.00&w12sc & 0.014&48&@xmath74&@xmath75&@xmath76&@xmath77&@xmath78&@xmath79&@xmath80 + -1.50&w16sp & 0.016&75&@xmath81&@xmath82&@xmath83&@xmath84&@xmath85&@xmath86&@xmath87 + -1.63&k16sp & 0.019&77&@xmath88&@xmath89&@xmath90&@xmath91&&&@xmath92 + -2.00&w16sp & 0.027&37&@xmath93&@xmath94&@xmath63&@xmath95&@xmath96&@xmath97&@xmath98 + -2.23&k16sp & 0.027&12&@xmath99&@xmath100&@xmath101 & @xmath102&&&@xmath103 + | we have measured an extensive set of @xmath0 colors for m31 globular clusters @xcite .
we compare the predicted simple stellar population colors of three population synthesis models to the intrinsic colors of galactic and m31 globular clusters .
the best - fitting models fit the cluster colors very well the weighted mean color offsets are all @xmath1 mag .
the most significant offsets between model and data are in the @xmath2 and @xmath3 passbands ; these are not unexpected and are likely due to problems with the spectral libraries used by the models .
the metal - rich clusters ( @xmath4}\gtrsim-0.8 $ ] ) are best fit by young ( 8 gyr ) models , while the metal - poor clusters are best fit by older ( 1216 gyr ) models .
if this range of globular cluster ages is correct , it implies that conditions for cluster formation must have existed for a substantial fraction of the galaxies lifetimes . |
recent observations have revealed the existence of a population of hypervelocity stars ( hvss ) traveling in the halo of the milky way ( mw ) with galactic rest - frame velocities @xmath2 in the range between @xmath3 and @xmath4 @xcite .
hvss are probably b - type main sequence stars with lifetimes @xmath5 , galactocentric distances @xmath6 kpc , and move with speeds large enough to escape from the galaxy .
the significant excess of b - type stars with velocities @xmath7 and distances @xmath8 kpc observed by @xcite may also be an indication that many hvss are ejected into the halo on _ bound _ orbits .
hvss were first recognized by @xcite as an unavoidable byproduct of the presence a massive black hole ( bh ) in the galactic center .
only a close encounter with a relativistic potential well can accelerate a 3 - 4 @xmath9 star to such extreme velocities , and at least three different ejection mechanisms have been proposed : the interaction between background stars and an intermediate - mass black hole ( imbh ) inspiralling towards sgr a@xmath10 @xcite , the disruption of stellar binaries in the tidal field of sgr a@xmath10 @xcite , and the scattering of stars off a cluster of stellar - mass bhs orbiting sgr a@xmath10 @xcite in all these models , hvss have unique kinematics compared to other halo stars : 1 ) they have almost zero initial specific angular momentum at ejection , @xmath11 , where @xmath12 the mass of sgr @xmath13 and @xmath14 the pericenter distance of the star ; 2 ) their high speeds diminish the impact of two - body relaxation or dynamical friction effects on their motion ; and 3 ) their trajectories will deviate from being exactly radial because of the asymmetry of the milky way potential produced by the flattened disk and the triaxial dark matter ( dm ) halo , causing a change of angular momentum that can be much larger than the initial small value .
( for reference , a @xmath15 deviation of the velocity from the radial direction at @xmath16 represents a change of @xmath17 in specific angular momentum . ) proper - motion measurements of hvss may therefore become a key diagnostic tool for constraining the shape of the galactic potential @xcite .
triaxial halos are a generic prediction of the hierarchical , cold dark matter ( cdm ) models of structure formation .
dissipationless cosmological simulations typically predict minor - to - major density axis ratios in the range 0.4 - 0.8 ( e.g. @xcite ) , with the asphericity of the potential increasing rapidly towards the center of the halo @xcite .
gas cooling tends to circularize the potential ( e.g. @xcite ) , while subsequent mergers produce highly elongated remnants ( e.g. @xcite ) .
studies of weak gravitational lensing and x - ray observations of elliptical galaxies show that halos are significantly flattened , in fair agreement with results from numerical simulations @xcite . yet
the coherence of tidal debris from the sagittarius dwarf galaxy appears to indicate that the inner halo of the mw is nearly spherical and therefore in conflict with cdm predictions ( ibata et al .
2001 ; but see helmi 2004 ) . in this paper
, we study the kinematics of hvss in the mw as a probe of the triaxiality of the galactic halo .
the outline is as follows . in [ sec : analysis ] , we analyze the motion of hvss in a flattened or triaxial gravitational potential .
we provide a concise statistical estimator for the triaxiality of the galactic halo potential through the measured angular momenta of hvss . in [ sec : potential ] , we review the galactic potential model to be used in our calculations . in [ sec : simulation ] we perform numerical simulations of the motion of hvss to study their kinematics . finally , in [
sec : conclusion ] , we summarize our conclusions .
consider a star with position vector @xmath18 moving with velocity @xmath19 in a gravitational potential @xmath20 , where @xmath21 and @xmath22 are the spherically - symmetric and aspherical component of the the potential , @xmath23 are cartesian coordinates , and @xmath24 .
the rate of change of the specific angular momentum of the star , @xmath25 , is equal to the torque , d / dt=-=- , [ eq : djdt ] and has components [ eq : djidt ] it is convenient to change from cartesian to spherical coordinates , @xmath26 , and combine the above equations to yield + + = 0 .
[ eq : dj0 ] from the definition of angular momentum it is also easy to derive _ x+_y+_z=0 , [ eq : calj0 ] where are determined directly from the position and velocity of the star .
note that equations ( [ eq : dj0 ] ) and ( [ eq : calj0 ] ) are rotationally invariant , that is , they do not change when arbitrary rotations are applied to their arguments . below we apply the above analysis to the motion of stars in two simple cases of non - spherical potentials . *
if the non - spherical component of the gravitational potential is axisymmetric about the plane @xmath27 , = ( r=,z ) , [ eq : phiaxis ] then @xmath28 , and @xmath29 is conserved .
stars ejected from the galactic center on radial orbits move in a plane with _ x=-_y , _
[ eq : caljaxisy ] * if the non - spherical component of the potential is triaxial , = ( x^2+y^2/p^2+z^2/q^2 ) , then a triaxiality parameter can be defined as t. [ eq : tdef ] if @xmath30 , the potential reduces to the spherical case .
if @xmath31 and @xmath32 ( @xmath33 ) , @xmath34 and @xmath35 , or @xmath36 ( @xmath37 ) , the potential is axisymmetric .
if @xmath38 , the triaxiality parameter is @xmath39 . in a triaxial potential , equation ( [ eq : djidt ] )
can be written as = -t .
[ eq : tdj ] for hvss moving away from the galactic center on radial orbits , the deviation of their trajectory from the initial ejection direction , ( @xmath40 ) , is small . replacing the angles @xmath41 in equation ( [ eq : tdj ] ) with @xmath42 and integrating yields t & = & -[1+++ + & & + ... ] , [ eq : tcalj ] where
the @xmath43 and @xmath44 arguments have been kept constant in the integration .
the term in parenthesis specifies the maximum systematic error on the triaxiality estimator @xmath45 , and numerical calculations ( see 4 below ) of the motion of hvss in a triaxial potential show that the typical error on @xmath46 is actually smaller . for a sample of @xmath47 hvss we can use equation ( [ eq : tcalj ] ) to construct a statistical estimator of triaxiality as [ eq : terror ] here t_i-=- , and
@xmath48 is its error .
note that the parameters @xmath49 are fully determined by the instantaneous positions and velocity vectors of the hvss in the sample , and that , while the ratio @xmath50 can also be used to estimate the triaxiality @xmath51 , this does not provide any independent information since it can be trivially derived from @xmath52 using equation ( [ eq : calj0 ] ) .
+ in the following we set the @xmath53-axis of the triaxial potential to be normal to the galactic disk plane , and denote with @xmath54 the angle measured counter - clockwise from a reference direction ( e.g. the line from the galactic center to the sun ) to the @xmath55-axis of the potential .
the ratio @xmath56 in a frame forming an angle @xmath57 with the reference direction can be written as -= , where a=,b=. if @xmath58 , the values @xmath56 may spread out over a large range due to the different angles @xmath59 in the sample .
the angle @xmath54 can be estimated by minimizing the weighted variance of @xmath60 : _
i=1^n()^2 , [ eq : variance ] where @xmath61 is the error of @xmath62 and the values of @xmath63 and @xmath64 are obtained from equations ( [ eq : tbar ] ) and ( [ eq : terror ] ) .
we use here a four - component model for the gravitational potential of the milky way , @xmath65 , where ( cf . @xcite ) : * @xmath66 is the contribution of sgr a@xmath10 , = - , [ eq : phibh ] with mass @xmath67 @xcite .
the radius of influence of sgr a@xmath10 is @xmath68 , where @xmath69 is the one - dimensional stellar velocity dispersion in the galactic center .
* @xmath70 is the contribution of the spherical bulge @xcite , = - , [ eq : phib ] with mass @xmath71 and core radius @xmath72 .
as both the bulge mass and size are small compared to those of the galactic disk and halo , a slight deviation of the bulge from sphericity will not have a significant effect on the change of angular momentum of hvss in the halo .
* @xmath73 is the contribution of the axisymmetric disk @xcite , ( r , z)=- , [ eq : phid ] with mass @xmath74 , scale length @xmath75 , and scale height @xmath76 .
* @xmath77 is the contribution of the triaxial dark matter halo , @xmath78 here @xmath79 is the halo concentration parameter , @xmath80 the radius within which the enclosed average density is 200 times the mean matter density , @xmath81 the scale radius , and @xmath82 .
this generalization of an nfw @xcite model ensures that the spherically - averaged potential of the triaxial halo is similar to that of a spherical halo with the same mass @xmath83 and scale radius @xcite .
we choose the following parameters for the milky way halo : @xmath84 , @xmath85 , @xmath86 ( e.g. @xcite ) , and assume that the @xmath53-axis of the halo and disk potentials coincide .
we also assume , for simplicity , that the asphericity of the potential is constant with radius @xmath87 , and set @xmath88 and @xmath89 in the calculations below .
cdm halos are not smooth but have a wealth of substructure on all resolved mass scales ( e.g. @xcite ) .
the assumption made above of a smooth gravitational potential neglects the deflection of hvs trajectories by halo substructure .
a star with velocity @xmath90 passing within a distance @xmath91 from a subhalo of mass @xmath92 will change its velocity by an amount @xmath93 .
the mass function of substructure in a milky way - sized halo can be described as @xmath94 in the subhalo mass range @xmath95 @xcite , with @xmath96 .
the probability that a hvs ejected from the galactic center is subject to a velocity deflection @xmath97 can then be estimated as @xmath98 where @xmath99 is the halo circular velocity .
this probability is quite low so that a smooth potential is a good assumption in this work .
in this section we perform numerical calculations of the motion of hvss in the mw gravitational potential . according to the study of @xcite , three - body interactions between ambient stars and a bh pair ( where the secondary bh may be an imbh inspiralling towards sgr a@xmath10 ; e.g. , @xcite ) expel hvss ( @xmath100 ) at a rate that can be as large as @xmath101 ( for a binary with semimajor axis @xmath102 and mass ratio of 0.01 ) .
tidal break - up of binary stars ( `` hills mechanism '' ) ejects hvss at a rate @xmath103 , where @xmath57 is the fraction of stars in binaries with semimajor axis @xmath104 .
close encounters of two single stars in the vicinity of sgr a@xmath10 may also produce hvss but at the negligible rate of @xmath105 .
different ejection mechanisms give origin to different spatial and velocity distributions of hvss .
while hills mechanism predict hvss to be expelled isotropically at an approximately constant rate , in models involving a bh pair hvss are ejected preferentially within the orbital plane of the binary in a short burst lasting a few myr @xcite . in the latter case
the degree of anisotropy depends on binary separation , the mass ratio , and the orbital eccentricity of the bh binary . for illustrative purposes , we assume in this paper that hvss
are ejected from the galactic center by the bh binary mechanism .
we use the stellar spatial and velocity distributions derived from the scattering experiments of @xcite for a binary with mass ratio @xmath106 , semimajor axis @xmath107 , and eccentricity @xmath108 .
the binary orbit is in the galactic disk plane , and the velocity of the lighter hole at pericenter is directed along @xmath109 .
the `` hardening '' radius @xmath110 is defined as @xcite @xmath111 when @xmath112 , the binary separation decreases both by dynamical friction and three - body interactions with low - angular momentum stars passing in its immediate vicinity .
after the binary becomes `` hard '' ( @xmath113 ) , the bound pair loses orbital energy mainly through three - body interactions until gravitational radiation takes over @xcite .
the ejection speed of the stars at infinity spans a range with r.m.s .
@xmath114 ( see eq . 33 in @xcite ) . in the calculations below
, we assume @xmath115 hvss are ejected from the galactic center at a constant rate over the last @xmath116 years , and ignore for simplicity the orbital evolution of the binary during such timescale .
different stars move independently in the galactic potential . a fourth - order runge - kutta method with adaptive stepsize control
is used to solve the differential equations of the motion of the stars .
note that the adopted mass ratio and semimajor axis are within the allowed parameter space for a bh pair at the galactic center ( see fig . 2 in @xcite ) . , and its present - day velocity and galactocentric distance are @xmath117 and @xmath118 , respectively . ]
( a ) ( _ solid line _ ) and ( e ) ( _ dotted line _ ) with @xmath119 and @xmath120 .
the bound stars are clustered around @xmath121 and @xmath122 in a spherical halo potential ( _ dotted line _ ) , while their distribution is spread uniformly over the range @xmath123@xmath122 in a triaxial halo potential ( _ solid line _ ) . ]
( a ) with @xmath124 and @xmath120 . _ bottom panel : _ all stars shown in fig .
[ fig : alphar](b ) with @xmath125 and @xmath126 .
hvss with short travel times , @xmath127 myr ( _ top _ ) and @xmath128 ( _ bottom _ ) have @xmath129 angles smaller than @xmath130 . the large cross in the top panel flags the locus of the bound star whose orbit is shown in fig .
[ fig : orbit ] . ]
we use the initial conditions described above and numerically integrate the orbits of hvss in the galactic potential . figures [ fig : map ] and [ fig : alphar ] show maps of stellar position and velocity vectors at the present time and their deflection angles @xmath131 . for hvss with @xmath132 , deviations are quite small , @xmath133 , at all distances within @xmath134 . lower velocity stars at small distances can instead be bound , and their deflection angles extend to @xmath122 .
many of the stars with @xmath135 and @xmath136 in figure [ fig : alphar](a ) follow bound trajectories , while no stars in this velocity range have substantial deviation angles at large galactocentric distances .
one example of a bound orbit is shown in figure [ fig : orbit ] , where the trajectory has been significantly bent by the triaxial halo and the flattened disk potentials , and the star does not return to the galactic center .
note that bound stars with @xmath137 and large @xmath129 s are typically observed at smaller distances ( @xmath138 ) than bound stars of lower velocities .
this is because a larger ejection speed from the galactic center is needed to maintain a high velocity at large distances , and such stars may then have either escaped from the halo or not have had sufficient time to come back and show a significant bend in their orbits .
( for reference , the local escape speed is about @xmath139 , @xcite . )
we find that 60% of the stars with @xmath140 at @xmath119 have velocities @xmath141 , and 20% of the stars with @xmath140 at @xmath142 have velocities @xmath143 .
figure [ fig : alphat ] shows that bound stars with large deflection angles ( @xmath144 have generally traveled a long time after ejection ( @xmath14550 myr for @xmath136 ) , and many of them have experienced at least one orbital period ( fig .
[ fig : orbit ] ) .
hvss with small deflections ( @xmath146 ) have instead a short travel time and are generally on the initial phases of their first orbital periods ( see the concentration of the stars at the left bottom of the panels in figure [ fig : alphat ] ) . the transverse velocities ( in the galactocentric frame ) of stars with @xmath140 are typically higher than @xmath147 , and can be up to @xmath148 ( hundreds of @xmath149 ) for unbound ( bound ) hvss .
note that @xmath147 corresponds at a distance of 100 kpc to a proper motion of @xmath150as in three years , which can be resolved by the next generation of astrometric surveys like gaia .
according to our calculations most hvss would have transverse velocities @xmath151 even in the case of a weakly triaxial halo with @xmath152 : these transverse velocities are larger than those associated with the bending of stellar trajectories caused by the axisymmetric disk ( see fig . [
fig : alphar]h ) . a comparison between figures [ fig : alphar](e)(g ) and [ fig : alphar](a)(c ) shows that the distribution of deflection angles versus distance is different for trajectories in a spherical rather than triaxial halo . in figure
[ fig : alphar](e ) and ( f ) , the @xmath129 angles of bound stars are clustered around @xmath121 and @xmath122 because their orbits are highly eccentric with little bending ( see also fig .
[ fig : histogram ] ) .
by contrast , in figure [ fig : alphar](a ) and ( b ) the deflection angles of bound stars lies at intermediate values .
this difference can be used as an indicator of the triaxiality of the mw dark matter halo .
figure [ fig : mapjn](a)(c ) depicts the values of @xmath153 and @xmath154 versus distance of all the hvss plotted in figure [ fig : map ] having @xmath140 and far from the plane of the disk , i.e. with @xmath155 .
figure [ fig : mapjn](d ) shows all stars with @xmath156 in the range @xmath121@xmath122 . as seen in figure [ fig : mapjn](a ) , the ratio @xmath153 at large distances ( @xmath157 ) is close to the triaxiality of the halo potential ( @xmath158 , _ solid line _ ) assumed in our calculations , with only a small scatter . the scatter is larger in the quantity @xmath154 plotted in figure [ fig : mapjn](b ) .
the slight offset ( @xmath159 ) of the stellar dots from the solid line is partly due to the disk potential that causes an additional bending of stellar trajectories towards the disk plane , increasing @xmath160 without changing @xmath161 ( see the appendix for details about the correction of such offset owing to the galactic disk ) .
as shown in figure [ fig : mapjn](e ) , hvss become better tracers of triaxiality after removing the disk potential from our calculations .
the higher the velocities , the smaller the offset .
the curvature of the ratio @xmath153 towards smaller values at small distances is also an effect of the disk .
the ratio @xmath153 for stars close to disk plane is not a good approximation of halo triaxiality even at large distances ( see fig .
[ fig : mapjn]d ) .
we plot the values of @xmath161 versus @xmath162 in figure [ fig : mapjns ] , where panel ( a ) shows stars with @xmath163 , and panel ( b ) stars with @xmath156 in the full range @xmath121@xmath122 .
the effect of the disk potential can be seen in the spreading of the stellar dots from the solid line @xmath164 to the dashed line @xmath165 .
note that the scatter in the ratio @xmath166 increases significantly if the reference frame is different from the frame of the triaxial halo potential ( fig .
[ fig : mapjn]f ) .
we have studied the unique kinematics of hvss ejected from the galactic center with almost zero initial specific angular momentum .
hvss can travel in the galactic halo on either bound or unbound orbits , and their spatial and velocity distribution at large galactocentric distances ( @xmath167 ) contain information on the asphericity of the halo gravitational potential .
we have proposed an estimator of the triaxiality of the galactic dark matter halo that is determined solely by instantaneous position and velocity vectors of hvss , is independent of the details of the ejection mechanism , and does not require an accurate knowledge of halo mass .
future astrometric and deep wide - field surveys of hvss should detect significant numbers of hvss , which could be used to determine the triaxiality of the mw halo by applying the method proposed in this paper .
the new class of possibly bound hvss with velocities @xmath168 recently observed by @xcite has galactocentric distances in the range 30 - 60@xmath169 or 10 - 20@xmath169 depending on whether they are main - sequence or blue horizontal branch stars . in the first case ( main - sequence stars at large distances )
, they have a lifetime of @xmath170 and are , according to figure [ fig : alphat ] , on the initial phases of their first orbital periods .
their deflection angles are expected to be rather small , supporting the fact that a significant excess of b - type stars is observed only at large positive velocities @xcite .
if bound hvss are blue horizontal branch stars instead at smaller distances , the travel time of the stars can be much longer than @xmath171 as at the ejection moment the stars may not necessarily be blue horizontal branch stars but at some pre - blue - horizontal - branch stage .
such stars may have experienced at least one orbital period , and many of them should be returning to the galactic center or their orbits should have been significantly bent by the asymmetric galactic potential ( fig .
[ fig : alphat ] ) .
this scenario appear unlikely since it does not agree with the observed positive radial velocities .
it is interesting at this stage to provide an example of a statistical estimate of halo triaxiality from a mock sample .
let us assign a measurement error of @xmath172 ( @xmath173 ) to the one - dimensional velocities of all stars in figure [ fig : mapjn](a ) having @xmath140 , @xmath155 , and @xmath174 ( our error analysis assumes that the distances to hvss are known to within ten percent ) .
we have simulated the observed velocities of such a sample , and plot in the top panel of figure [ fig : texample ] the values of @xmath49 and @xmath175 derived for each hvs . using equations ( [ eq : tbar ] ) and ( [ eq : terror ] ) , we obtain @xmath176 and @xmath177 . for @xmath178 , we obtain @xmath176 and @xmath179 .
these errors are comparable with the systematic error caused by the flattened disk ( see the slight offset of dots from the solid line in fig . [
fig : mapjn]a ) . we have tried different values of the gravitational potential parameters @xmath180 in the calculations , and found that the estimated value of @xmath181 is always consistent with the assumed halo triaxiality ( see middle and bottom panels in fig .
[ fig : texample ] ) . , @xmath155 , and @xmath182 .
an observational error of @xmath183 is assumed for in `` measured '' one - dimensional velocity . from top to bottom , the values of @xmath180 used in the calculations are ( 0.8 , 0.7 ) , ( 0.8 , 0.6 ) , and ( 0.8 , 0.75 ) , corresponding to a triaxiality @xmath46=0.54 , 0.32 , and 0.72 , respectively ( _ horizontal lines _ ) . a statistical analysis of the sample using eqs . ( [ eq : tbar ] ) and ( [ eq : terror ] ) yields @xmath184=(0.50 , 0.02 ) , ( 0.31 , 0.01 ) , and ( 0.64 , 0.02 ) , respectively . for clarity ,
only 1/3 of the sample points are drawn in the figure . ] of the simulated sample ( top panel in fig .
[ fig : texample ] ) in different reference frames as a function of @xmath185 . here
@xmath54 is the angle measured counter - clockwise from a reference direction ( e.g. the line from the galactic center to the sun ) to the @xmath55-axis of the halo potential , and @xmath57 is the angle formed by the observational frame with the reference direction .
the variance has a minimum at @xmath186 . ] if the @xmath55-axis of the halo potential form an angle @xmath54 ( measured counter - clockwise ) with the reference axis of the observations , then ( as seen from fig . [ fig : variance ] ) the minimum variance of @xmath187 in a set of simulated samples each in a frame at angle @xmath57 from the observational reference axis ( see eq .
[ eq : variance ] ) occurs for @xmath188 .
our calculations show that an error of @xmath130 in the estimate of @xmath54 may cause an error of 2% in the estimate of @xmath181 .
note that the axis ratios @xmath180 of the halo potential are degenerate in the defined triaxiality parameter ( eq . [ eq : tdef ] ) . after determining @xmath46 ,
the values of @xmath180 could be also obtained by using any value of @xmath189 ( e.g. , see fig .
[ fig : mapjn]b ) , but the modeling would be sensitive to halo mass and the shape of halo potential used ( see also the determination of axis ratios in @xcite by tracing back hvs orbits ) . finally , it is possible that a few hvss in the halo may be produced by the interactions of stars with an imbh in satellite galaxies like the large magellanic cloud @xcite , and that these would contaminate the sample ejected from the galactic center .
the ejection rates from the satellite dwarfs of the mw are expected to be much smaller than the rate from sgr a@xmath10 , however .
such `` satellite '' hvss will also have much larger angular momenta in the galactocentric rest - frame , and should be easily distinguishable from the galactic center sample .
we have benefited from discussions with francesco haardt , youjun lu , and scott tremaine .
we thank alberto sesana for providing the initial velocity and spatial distributions of hvss ejected from a black hole binary .
p.m. acknowledges financial support from nasa through grants nag5 - 11513 and nng04gk85 g , and from the alexander von humboldt foundation . q.y . acknowledges initial support from nasa through hubble fellowship grant hst - hf-01169.01-a awarded by the space telescope science institute , which is operated by the association of universities for research in astronomy , inc .
, for nasa , under contract nas 5 - 26555 .
as shown in figure [ fig : mapjn](a ) , the triaxiality obtained through @xmath45 has a small offset from the true halo triaxiality .
part of this offset comes from the effects of the flattened disk potential , and part from the fixing of @xmath41 in the integration of equation ( [ eq : tdj ] ) .
the offset due to the constant @xmath41 assumption is smaller for stars of higher velocities , as seen in figure [ fig : mapjn](e ) that shows the results after the disk potential was removed .
below we provide a method to correct the offset due to the galactic disk . as in the determination of halo triaxiality proposed in this work , even for this correction
it is not necessary to trace back the orbits of hvss .
the contribution to the change in specific angular momentum due to the disk potential can be expressed as : j_y,=^r_0_r_ejdr , where = 2xz(- ) , [ eq : djdisk ] = , [ eq : vr ] @xmath190 is the initial distance of the hvs from the galactic center at ejection , @xmath191 and @xmath192 are the current position and velocity of the hvs , @xmath193 , and @xmath194 is its radial velocity at @xmath87 .
we can remove the effect of the disk potential from the variable @xmath162 by computing @xmath195 , where @xmath196 and where equation ( [ eq : phid ] ) describing the disk potential has been used .
the angles @xmath41 are fixed in the integration to be present - day values .
in addition to the disk potential , we also need to assume a halo potential to determine @xmath194 , for which we may use the spherical part of the halo potential by setting @xmath197 .
we have tested this correction and found that half of the offset to the true halo triaxiality can be corrected . | hypervelocity stars ( hvss ) ejected by the massive black hole at the galactic center have unique kinematic properties compared to other halo stars .
their trajectories will deviate from being exactly radial because of the asymmetry of the milky way potential produced by the flattened disk and the triaxial dark matter halo , causing a change of angular momentum that can be much larger than the initial small value at injection .
we study the kinematics of hvss and propose an estimator of dark halo triaxiality that is determined only by instantaneous position and velocity vectors of hvss at large galactocentric distances ( @xmath0 ) .
we show that , in the case of a substantially triaxial halo , the distribution of deflection angles ( the angle between the stellar position and velocity vector ) for hvss on bound orbits is spread uniformly over the range 10@xmath1180@xmath1 .
future astrometric and deep wide - field surveys should measure the positions and velocities of a significant number of hvss , and provide useful constraints on the shape of the galactic dark matter halo .
[ firstpage ] black hole physics galaxy : center galaxy : halo stellar dynamics |
this work is supported by the swedish research council ( vr ) under grant nos . 621 - 2012 - 3805 , and 621 - 2013 - 4323 and the gran gustafsson foundation .
we also thank d.s .
delion for discussions and for his efforts studying above - mentioned nuclei within the coherent state model and p. maris for his effort calculating those nuclei with the code mfdn .
the computations were performed on resources provided by the swedish national infrastructure for computing ( snic ) at pdc , kth , stockholm .
62ifxundefined [ 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 * * , ( ) http://www.sciencedirect.com/science/article/pii/s0370269310012864 [ * * , ( ) ] @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop http://cds.cern.ch/record/1981273[__ ] , ( , , ) @noop * * ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop http://www.nndc.bnl.gov/nudat2/ [ `` , '' ] @noop * * , ( ) @noop * * , ( ) @noop _ _ ( , ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) | we present systematic calculations on the spectroscopy and transition properties of even - even te isotopes by using the large - scale configuration interaction shell model approach with a realistic interaction .
these nuclei are of particular interest since their yrast spectra show a vibrational - like equally - spaced pattern but the few known e2 transitions show anomalous rotational - like behavior , which can not be reproduced by collective models .
our calculations reproduce well the equally - spaced spectra of those isotopes as well as the constant behavior of the @xmath0 values in @xmath1te .
the calculated @xmath0 values for neutron - deficient and heavier te isotopes show contrasting different behaviors along the yrast line .
the @xmath0 of light isotopes can exhibit a nearly constant bevavior upto high spins .
we show that this is related to the enhanced neutron - proton correlation when approaching @xmath2 .
the advent of large - scale radioactive beam facilities and new detection technologies have enabled the study of the spectroscopy and transition properties of @xmath3 nuclei just above the presumed doubly magic nucleus @xmath4sn @xcite .
several unexpected phenomena have recently been observed : large @xmath5 values of neutron deficient semi - magic @xmath6 isotopes have triggered extensive experimental @xcite and theoretical @xcite activities , in particular regarding the fundamental roles played by core excitations and the nuclear pairing correlation ( or seniority coupling ) .
the study of transition rates in isotopic chains just above @xmath7 may provide further information on the role of core excitations @xcite .
the limited number of valence protons and neutrons are not expected to induce any significant quadrupole correlation in this region @xcite .
the low - lying collective excitations of those nuclei were discussed in terms of quadrupole vibrations @xcite in relation to the fact that the even - even te isotopes between @xmath8 and 70 show regular equally - spaced yrast spectra ( c.f . , fig . 1 in ref .
@xcite ) .
if that is the case , the te isotopes will provide an ideal ground to explore the nature of the elusive nuclear vibration and the residual interactions that leading to that collectivity .
however , the available e2 transition strengths along the yrast line in @xmath9te show an anomalous rotational - like behavior , which can not be reproduced by collective models or the interacting boson model @xcite .
another intriguing phenomenon is the nearly constant behavior of the energies of the @xmath10 and @xmath11 states in te and xe isotopes and their ratios when approaching @xmath2 , in contrast to the decreasing behavior when approaching @xmath12 @xcite .
this was analyzed in ref .
@xcite based on the quasiparticle random phase approximation approach where an competition between the quadurople - quadrupole correlation and neutron - proton pairing correlation was suggested .
an enhanced interplay between neutrons and protons is expected in the @xmath4sn region since the protons and neutrons partially occupy the same quantum orbitals near the fermi level @xcite . in relation to that , there has also been a long effort searching for superallowed alpha decays from those @xmath3 isotopes @xcite .
the region is also expected to be the endpoint of the astrophysical rapid proton capture ( rp ) process @xcite .
the octupole correlation may also play a role here ( the coupling between the @xmath13 and @xmath14 orbitals ) @xcite .
still , compared to tin , the experimental information is less abundant in the isotopic chain of tellurium where little was known experimentally below the neutron midshell until recently . much more work is needed and further measurements are underway in order to map out the ordering and nature of single - particle states and two - body effective interactions in the region @xcite . in this work
we present systematic large - scale calculations on the spectroscopy and transition properties of even - even te isotopes .
the large - scale shell model , which takes into account all degrees of freedom within a given model space , is an ideal approach to study competition between collective and single - particle degrees of freedom .
it is however a challenge , especially in the midshell , due to the huge dimension of the problem ( c.f . ,
fig . 1 in ref .
@xcite ) . on the theoretical side ,
we have done systematic calculations on the @xmath15 values of even - even te isotopes @xcite .
the results are , however , rather sensitive to the truncation imposed .
now we are able to do full shell model calculations for all low - lying states of all te isotopes with further optimization of the shell - model algorithm .
a full shell model calculation for the spectroscopy of @xmath16te was done in ref .
@xcite .
a schematic calculation for @xmath17te in the @xmath18 subspace was presented in ref .
@xcite .
systematic calculations on the sn and sb isotopes were given in refs . and
ref .
@xcite , respectively . in ref .
@xcite , the possible onset of vibrational collectivity in te isotopes was discussed within an effective field theory framework .
we consider the neutron and proton orbitals between the shell closures @xmath19 ( and @xmath20 ) @xmath21 and 82 , comprising @xmath22 , @xmath14 , @xmath23 , @xmath24 and @xmath13 and assume @xmath4sn as the inert core .
the robustness of the @xmath25 shell closures , which has fundamental influence on our understanding of the structure of nuclei in this region , is supported by recent measurements @xcite .
the nearly degenerate neutron single - particle states @xmath26 and @xmath27 orbitals in @xmath28sn were observed by studying the @xmath29-decay @xmath30te @xmath31 @xmath28sn @xcite .
based on the assumption that the ground state of @xmath30te has spin - parity @xmath32 , the @xmath27 orbital was suggested to be the ground state of @xmath28sn instead of @xmath26 .
the excitation energy of the @xmath14 is taken as @xmath33 mev .
the energies of other states have not been measured yet .
they are adjusted to fit the experimental binding energies of tin isotopes .
the starting point of our calculation is the realistic cd - bonn nucleon - nucleon potential @xcite .
the interaction was renormalized using the perturbative g - matrix approach to take into account the core - polarization effects @xcite .
the @xmath34 part of the monopole interaction was optimized by fitting to the low - lying states in sn isotopes @xcite .
further optimization of the @xmath35 part of the interaction is also underway which , however , is still a very challenging task .
our calculations show that the present effective hamiltonian are already pretty successful in describing the structure and transition properties of sb , te , i , xe and cs isotopes as well as heavy nuclei near @xmath12 in this region .
the te isotopic chain is the heaviest and longest chain that can be described by the nuclear shell model .
the dimension for the mid - shell @xmath36te reaches @xmath37 for which the diagonalization is still a very challenging numeric task . in our previous calculations for the @xmath15 of mid - shell te isotopes @xcite , we restricted a maximum of four neutrons that can be excited from below the fermi surface to the neutron @xmath38 subshell and excluded proton excitation to @xmath38 due to limited computation power , which , as we understand now , is a rather severe truncation .
the model space was further extended to allow at most 8 particles to the @xmath38 subshell in ref .
@xcite , where the @xmath15 values show a much smoother parabolic behavior as a function of @xmath19 .
full shell - model calculations are done for all nuclei in the present work .
all shell - model calculations are carried out within the @xmath39-scheme where states with @xmath40 are considered .
diagonalizations are done with a parallel shell model program that we developed @xcite and with a slightly modified version of the code kshell @xcite .
all calculations are done on the supercomputers beskow and tegnr at pdc center for high performance computing at the kth royal institute of technology in stockholm , sweden . to test the validity of the effective interaction ,
we have done systematic calculations on the yrast spectra of isotopes @xmath41te .
the results for @xmath42te are plotted in fig .
[ sm ] in comparison with available experimental data @xcite .
an overall good agreement between theory and experiment is obtained .
noticeable difference is seen in the excitation energies of the @xmath43 states in @xmath44te and the @xmath45 states in @xmath17te .
all isotopes plotted in the figure show rather regular and vibrational - like spectra up to @xmath43 except @xmath17te . for that nucleus , the calculated spectrum still shows a equally - spaced pattern .
however , a smaller gap between @xmath46 and @xmath47 states is expected from recent measurement but the spin - parity assignments for those states are still tentative .
the equally - spaced pattern breaks down in isotopes heavier than @xmath48te where a gradual depression of the excitation energies of the @xmath49 states is seen .
comparison between theory and experiment for the energies of the @xmath10 and @xmath11 yrast states ( upper ) and the @xmath15 values ( lower ) for the te isotopic chain .
the open circles and open triangles in the lower panel correspond to the square of the neutron and proton matrix elements , @xmath50 and @xmath51 , respectively.,title="fig : " ] comparison between theory and experiment for the energies of the @xmath10 and @xmath11 yrast states ( upper ) and the @xmath15 values ( lower ) for the te isotopic chain .
the open circles and open triangles in the lower panel correspond to the square of the neutron and proton matrix elements , @xmath50 and @xmath51 , respectively.,title="fig : " ] a closer comparison between experiment and calculation on the excitation energies of the yrast @xmath10 and @xmath11 states are plotted in fig .
[ sm2 ] as a function of neutron number for all even - even te isotopes . in the lower panel of the figure , the calculated @xmath15 in tellurium isotopes
are compared to the most recent experimental data @xcite .
the b(e2 ) value is calculated as @xmath0 = @xmath52 where @xmath53 and @xmath54 are the proton and neutron matrix elements and we take effective charges @xmath55 = 1.5e and @xmath56 = 0.8e as were employed in @xcite .
the isospin dependence of the effective charges is not considered here , which is not expected to have large influence on the trend .
the model prediction agrees rather well with available data .
the largest deviations are seen in isotopes @xmath57te .
a recent measurement gave a value smaller than the adapted one in the former case .
in the figure we also plotted the square of the neutron and proton matrix elements , @xmath50 and @xmath51 , which represent the separate contributions to the @xmath0 values from the neutron and proton excitations . as can be seen from the figure , the parabolic behavior of the @xmath0 values , which looks similar to that of sn isotopes , is mostly due to the contribution from the neutron excitation .
the contribution from the proton excitation shows a rather smooth and slightly decreasing behavior as the neutron number increases .
as mentioned earlier , the shell - model calculations for mid - shell te isotopes , in particular @xmath36te , are quite sensitive to the filling of both the proton and neutron @xmath38 subshells .
both the proton and neutron transition matrix elements are enhanced when one goes from a small model space calculation with restricted number of particles in @xmath38 to the full shell - model calculation .
comparison between theory and experiment @xcite for @xmath58 values in @xmath1te along the yrast line .
the open circles and open triangles in the lower panel correspond to the square of the neutron and proton matrix elements , @xmath50 and @xmath51 , respectively .
the dashed lines correspond to the predictions of collective models @xcite . ]
the regularly - spaced level spectra in mid - shell te isotopes have been expected to be be associated with a collective vibrational motion . for a spectrum corresponds to a vibrator , there should be collective e2 transitions between states differing by one phonon .
the transition strengths should be linearly proportional to the spin of the initial states , i.e. , one has @xmath59=2 in the harmonic vibrator model .
unfortunately , there are very few data available for @xmath0 values in states beyond @xmath60 . as shown in ref .
@xcite , the measured @xmath0 values along the yrast line in @xmath1te show a rather anomalous constant behavior up to @xmath61 , which looks more like a rotor and is in contradiction with that for a vibrator .
our shell model calculations for those e2 transitions are shown in fig .
[ sm3 ] , which indeed exhibits a rather constant behavior up to higher spins .
moreover , both the proton and neutron matrix elements remain roughly the same along the yrast line . as can be seen fig .
[ sm3 ] , the ratio @xmath59 for @xmath1te is measured to be even slightly smaller than one .
this is not seen in the theory .
the ratio is calculated to be @xmath62 which actually agree well with the prediction for a rotor .
as discussed in refs .
@xcite , it happens rarely in open - shell nuclei that one has @xmath63 .
the reason why the ratio for @xmath1te is observed to be so small is still not clear .
calculated @xmath58 values along the yrast line for even - even te isotopes .
, title="fig : " ] calculated @xmath58 values along the yrast line for even - even te isotopes . ,
title="fig : " ] calculated @xmath58 values along the yrast line for even - even te isotopes .
, title="fig : " ] calculated @xmath58 values along the yrast line for even - even te isotopes .
, title="fig : " ] in ref .
@xcite , the ratios @xmath59 for isotopes @xmath64te are measured to be 1.640 , 1.500 and 1.162 , respectively .
the ratios calculated from the shell model @xmath0 values are 1.322 , 1.299 , and 1.301 , respectively , for above three nuclei , which agree reasonably with experimental data . in fig .
[ sm4 ] we plotted the calculated @xmath0 values for the yrast states of all even - even te isotopes between @xmath65 and 80 .
as can be seen from the upper panel of the figure , the @xmath0 values for the yrast states of @xmath66te , which are at the beginning of the shell , remain roughly constants up to spin @xmath67 and decrease significantly around @xmath68 , which indicates that the collectivity has collapsed there . on the other hand , as shown in the lower panel of the figure , the results for the isotopes @xmath69te at the end of the shell show a very different behavior : the @xmath0 values decrease dramatically after @xmath70 , which reach practically zero value for states up to @xmath71 in all nuclei except @xmath72te .
the results for the groups @xmath73 and @xmath74 are shown in the middle panels of fig .
[ sm4 ] . in the former group ,
the @xmath0 values show a rather constant behavior up to @xmath61 .
the results for @xmath71 diverge in relation to the fact that several low - lying @xmath75 states are predicted for those nuclei by the shell - model calculations and , in cases like @xmath76te shown in the panel , it is the second @xmath75 state that is connected to the yrast @xmath77 state with strong e2 transition . as a result ,
the @xmath78 value vanishes . in the latter group
, the @xmath0 values also show a large decrease after @xmath70 but to a extent that is much less than those in the fourth group , @xmath69te . to understand the behavior of the calculated @xmath0 values seen in fig .
[ sm4 ] , we notice that the ratio @xmath79 roughly equally to two for all known te isotopes below @xmath80 . but it decreases rapidly to around 1.2 in the semi - magic @xmath81te . moreover , the ratio @xmath82 starts to decrease already at @xmath83 , resulting in seniority - like spectra .
the seniority quantum number refers to the number of particles that are not paired to @xmath84 .
it is known that , for systems involving the same kind of particles , the low - lying states can be well described within the seniority scheme @xcite .
this is related to the fact that the @xmath34 two - body matrix elements is dominated by monopole pairing interactions with @xmath84 .
the seniority coupling may be broken by the neutron - proton correlation if both protons and neutrons are present .
this indeed happens in the most neutron deficient te isotopes close to @xmath85 , where the valence neutrons and protons are expected to occupy identical @xmath27 and @xmath26 orbitals and the neutron - proton correlation is expected to be strong . as a results , both the spectra and @xmath86 transition properties show rather regular collective behaviors . on the other hand , for nuclei @xmath69te at the other end of the shell ,
the normal seniority coupling may prevail since the neutron - proton correlation involves particles in different shells and is much weaker . as a result ,
the energy gap between @xmath49 and @xmath11 as well as the @xmath86 transition between the two states reduce significantly ( e.g. , @xmath86 transitions between states with the same seniority is disfavored ) .
the groups @xmath73 and @xmath74 fall between above two cases . if the dimension is not too large , it is possible to project the wave function as a coupling of the proton group and neutron group with good angular momenta in the form @xmath87 where @xmath88 and @xmath89 denote the angular momenta of the proton group and neutron neutron group ( see , e.g. , @xcite ) .
the ground state for a even - even nucleus will be represented by a single configuration with @xmath90 if there is no neutron - proton correlation .
the neutron - proton interaction induces contributions from configurations with higher angular momenta for the protons and neutrons as well as higher - lying configurations with the same angular momenta .
it is seen that , as expected , the @xmath16te ground state shows a high mixture of many component , among which one has 47% with @xmath90 , 30% with @xmath91 , 12.5% with @xmath92 , and 6.7% with @xmath93 . for @xmath17te ground state the results are 41.7% with @xmath90 , 42 % with @xmath91 , 11.9% with @xmath92 . for @xmath17te ground state
the contribution from @xmath90 decreases further to 36.3% while the contribution from @xmath91 increases to 46.6% .
the wave functions for other low - lying states show a similar complex structure . on the other hand , the wave functions for the low - lying states of the isotopes
@xmath69te show a much simpler picture and are dominated by either neutron or proton excitations in many cases .
the contributions from @xmath90 components are 65.2% , 74.4% and 85.5% for isotopes
@xmath94te . the @xmath60 state in @xmath95te
is dominated by @xmath96 whereas @xmath97 state is dominated by the proton excitation @xmath98 instead .
the low - lying states for @xmath99te show a similar result
. it may be interesting to mention that a similar picture with rotational - like @xmath0 transitions and vibrational - like spectrum is also predicted for @xmath85 nuclei like @xmath100pd @xcite in relation to the quest for the possible existence of np pairing in @xmath3 nuclei for which there is still no conclusive evidence after long and extensive studies ( see , recent discussions in refs .
@xcite ) .
moreover , the @xmath29 formation amplitude may increase as a result of of the strong neutron - proton correlation .
there has already been a long effort answering the question whether the formation probabilities of neutron - deficient @xmath3 te and xe isotopes are larger compared to those of other nuclei @xcite .
we have evaluated within the shell - model approach the @xmath29 formation amplitude @xcite .
if the neutron - proton correlation is switched on , in particular if a large number of levels is included , there can be indeed significant enhancement of @xmath29 formation amplitude . to summarize ,
we have done systematic calculations on the spectroscopy and transition properties of te isotopes within the large - scale configuration interaction shell model approach .
a monopole - optimized realistic interaction is used .
the calculations reproduce well the excitation energies of the low - lying states as well as the regular and vibrational - like behavior of the yrast specta of @xmath101te ( fig .
1 ) .
the energies of the first @xmath10 and @xmath11 states as well as their ratios show rather a rather constant behavior when approaching @xmath2 in relation to the enhanced neutron - proton correlation ( fig .
2 ) . on the other hand ,
a squeezed gap between the @xmath97 and @xmath47 states is expected when approaching @xmath12 , resulting in seniority - like spectra .
those structure changes are also reflected in the calculated and available experimental e2 transition strengths .
the calculated @xmath102 show a parabolic behavior as a function of @xmath19 , which is dominated by the contribution from the neutron excitation .
moreover , the calculations reproduced reasonably well the nearly constant behavior of the @xmath0 values of @xmath1te and @xmath64te along the yrast line ( figs . 3 & 4 ) . the anomalous constant behavior is related to the competition between the seniority coupling and the neutron - proton correlations . for neutron - deficient te isotopes , the constant behavior of @xmath0 values can be extended to high spin values around @xmath103 14 .
wheras for heavier isotopes , when the neutron - proton correlation gets weaker , the @xmath0 values can reduce significantly after @xmath70 and vanishes for the heaviest isotopes . |
observations of rotation curves of spiral galaxies and measurements of the velocity dispersions of stars in early - type galaxies have provided important evidence for the existence of massive dark matter halos around galaxies ( e.g. , van albada & sancisi 1986 ) .
in addition , these studies have presented evidence of tight relations between the baryonic and dark matter components ( e.g. , tully & fisher 1977 ; faber & jackson 1976 ) .
results based on strong lensing by galaxies support these findings ( e.g. , keeton , kochanek & falco 1998 ) .
the origin of these scaling relations must be closely related to the process of galaxy formation , but the details are still not well understood , mainly because of the complex behaviour of the baryons .
furthermore , on the small scales where baryons play such an important role , the accuracy of cosmological numerical simulations is limited .
this complicates a direct comparison of models of galaxy formation to observational data . for such applications , it would be more convenient to have observational constraints on quantities that are robust and easily extracted from numerical simulations .
an obvious choice is the virial mass of the galaxy , but most techniques for measuring mass require visible tracers of the potential , confining the measurements to relatively small radii .
fortunately , recent developments in weak gravitational lensing have made it possible to probe the ensemble averaged mass distribution around galaxies out to large projected distances .
the tidal gravitational field of the dark matter halo introduces small coherent distortions in the images of distant background galaxies , which can be easily detected in current large imaging surveys .
we note that one can only study ensemble averaged properties , because the weak lensing signal induced by an individual galaxy is too small to be detected . since the first detection of this so - called galaxy - galaxy lensing signal by brainerd et al .
( 1996 ) , the significance of the measurements has improved dramatically , thanks to new wide field ccd cameras on a number of mostly 4 m class telescopes .
this has allowed various groups to image large areas of the sky , yielding the large numbers of lenses and sources needed to measure the lensing signal .
for instance , hoekstra et al .
( 2004 ) used 45.5 deg@xmath11 of @xmath12-band imaging data from the red - sequence cluster survey ( rcs ) , enabling them to measure , for the first time , the extent and flattening of galaxy dark matter halos , providing strong support for the cold dark matter ( cdm ) paradigm .
however , the analysis presented in hoekstra et al . (
2004 ) was based on the @xmath12-band data alone , and consequently lacked redshift information for the individual lenses .
an obvious improvement is to obtain redshift information for the lenses ( and if possible the sources ) .
this allows one to study the lensing signal as a function of lens properties , most notably the luminosity .
photometric redshifts were used by hudson et al .
( 1998 ) to scale the lensing signal of galaxies in the hubble deep field , and by wilson et al .
( 2001 ) who measured the lensing signal around early - type galaxies as a function of redshift .
smith et al .
( 2001 ) and hoekstra et al .
( 2003 ) used spectroscopic redshifts , but the lens samples involved were rather small ( @xmath13 ) . the sloan digital sky survey ( sdss )
combines both survey area and redshift information .
its usefulness for galaxy - galaxy lensing was demonstrated clearly by fischer et al .
more recently , mckay et al .
( 2001 ) used the available sdss redshift information to study the galaxy - galaxy lensing signal as a function of galaxy properties ( also see guzik & seljak 2002 ; seljak 2002 ; sheldon et al .
2004 ) .
in this paper we use a subset of the rcs data , for which photometric redshifts have been determined using @xmath14 and @xmath15 data taken using the canada - france - hawaii telescope ( see hsieh et al . 2005 for details ) .
the area covered by these multiwavelength data is approximately 33.6 deg@xmath11 , resulting in a catalog of @xmath16 galaxies for which a redshift could be determined , making it one of the largest data sets of its kind .
this unique data set allows us to measure the virial masses of galaxies as a function of their luminosity .
this paper is structured as follows . in
2 we briefly discuss the data , including the photometric redshift catalog and its accuracy .
the results of some basic tests of the photometric redshifts are presented in 3 . in
4 we discuss the dark matter profile inferred from numerical simulations . the measurement of the virial mass as a function of luminosity in various filters is presented in 5 , as well as our measurement of the baryon fraction in galaxies . throughout the paper
we adopt a flat cosmology with @xmath17 , @xmath18 and a hubble parameter @xmath19 km / s / mpc .
the red - sequence cluster survey ( rcs ) is a galaxy cluster survey designed to provide a large sample of optically selected clusters of galaxies in a large volume ( see gladders & yee ( 2005 ) for a detailed discussion of the survey ) . to this end
, 92 deg@xmath11 of the sky were imaged in both @xmath12 and @xmath15 using the cfh12k camera on cfht and the mosaic ii camera on the ctio blanco telescope .
this choice of filters allows for the detection of clusters up to @xmath20 using the cluster red - sequence method developed by gladders & yee ( 2000 ) .
after completion of the original rcs survey , part of the surveyed area was imaged in both @xmath0 and @xmath1 band using the cfht .
this additional color information allows for a better selection of clusters at lower redshifts .
these follow - up observations cover @xmath21 deg@xmath11 , thus covering @xmath22 of the cfht fields .
the data and the photometric reduction are described in detail in hsieh et al .
( 2005 ) .
the galaxy - galaxy lensing results presented in hoekstra et al .
( 2004 ) were based on 45.5 deg@xmath11 of @xmath12-band data alone .
the addition of @xmath0 and @xmath1 imaging data for 33.6 deg@xmath11 to the existing @xmath12 and @xmath15 data allow for the determination of photometric redshifts for both lenses and sources in this subset of rcs imaging data .
this enables the study of the lensing signal as a function of the photometric properties of the lens galaxies ( i.e. , color and luminosity ) . in this paper
we focus on this multi - color subset of the rcs . to determine the restframe @xmath0 , @xmath1 and @xmath2 luminosities we use template spectra for a range in spectral types and compute the corresponding passband corrections as a function of redshift and galaxy color ( this procedure is similar to the one described in van dokkum & franx 1996 ) .
provided the observed filters straddle the redshifted filter of interest , which is the case here , this procedure yields very accurate corrections .
the cfht @xmath12 images are used to measure the shapes of galaxies used in the weak lensing analysis .
the raw galaxy shapes are corrected for the effects of the point spread function , as described in hoekstra et al .
( 2002a ) .
the resulting object catalogs have been used for a range of weak lensing studies ( e.g. , hoekstra et al .
2002a , 2002b , 2002c , 2004 ) and we refer to these papers for a detailed discussion of the shape measurements . the measurements of the lensing signal caused by large scale structure presented in hoekstra et al .
( 2002a , 2002b ) are very sensitive to residual systematics .
the various tests described in these papers suggest that the systematics are well under control . in this paper
we use the shape measurements to measure the galaxy - galaxy lensing signal , which is much less sensitive to these observational distortions : in galaxy - galaxy lensing one measures the lensing signal that is perpendicular to the lines connecting many lens - source pairs .
these are randomly oriented with respect to the psf anisotropy , and therefore residual systematics are suppressed .
the determination of the photometric redshifts is described in detail in hsieh et al .
the empirical quadratic polynomial fitting technique ( connolly et al . 1995 ) is used to estimate redshifts for the galaxies in the rcs data .
the key component in this approach is the creation of a training set .
spectroscopic redshifts from the cnoc2 survey ( yee et al .
2000 ) are matched to the corresponding objects in the overlapping rcs fields .
these data are augmented with observations of the goods / hdf - n field , for which the spectroscopic redshifts have been obtained using the keck telescope ( wirth et al .
2004 ; cowie et al .
2004 ) , and the photometry is from the ground - based hawaii hdf - n data obtained with the subaru telescope ( capak et al .
this results in a final training set that includes 4,924 objects covering a large range in redshifts . to minimize the fitting errors arising from different galaxy types , hsieh et al .
( 2005 ) used a kd - tree method with 32 cells in a three - dimensional color - color - magnitude space .
the resulting catalog contains @xmath16 galaxies with photometric redshifts .
this catalog was matched against the catalog of galaxies for which shapes were measured .
this resulted in a sample of @xmath23 galaxies with @xmath24 , that are used in the analysis presented here .
comparison with the spectroscopic redshifts shows that accurate photometric redshifts , with @xmath25 , can be derived in the range @xmath26 . at lower redshifts , the lack of @xmath27 band data limits the accuracy , whereas at higher redshifts photometric errors increase the scatter to @xmath28 ( see hsieh et al .
2005 for more details ) .
to study the halos of galaxies as a function of color and luminosity we select a sample of lenses at intermediate redshifts : we select galaxies with photometric redshifts @xmath3 and @xmath12-band magnitudes @xmath24 .
this redshift range is well covered by the cnoc2 redshift survey at the bright end , and the redshift errors are relatively small . for the background galaxies we limit the analysis to galaxies with @xmath29 .
figure [ dzlens ] shows the difference between spectroscopic and photometric redshifts for different subsets of galaxies with photometric redshifts @xmath3 .
panel a shows the full sample , whereas panels b and c show the bright and faint halves respectively .
the distribution is peaked , with 70% of the galaxies within the range @xmath30 ( 0.05 and 0.07 for the bright and faint subsets , resp . ) and 90% within @xmath31 ( 0.085 and 0.15 for the bright and faint subsets , resp . ) .
the solid histogram in figure [ zdist]a shows the normalized photometric redshift distribution for the galaxies brighter than @xmath32 .
it is common to parametrize the redshift distribution , and a useful form is given by @xmath33}\left(\frac{z}{z_s}\right)^\alpha \exp\left[-\left(\frac{z}{z_s}\right)^\beta\right].\ ] ] we fit this model to the observed redshift distribution .
however , the uncertainties in the photometric redshift determinations can be substantial , and as a result the observed distribution is broadened .
we use the observed error distribution , assuming a normal distribution , to account for the redshift errors . for the best fit parameterization
we find values of @xmath34 , @xmath35 ( fixed ) and @xmath36 , which yields a mean redshift of @xmath37 .
this model redshift distribution ( which includes the smoothing by redshift errors ) is indicated by the smooth curve in figure [ zdist]a . in the weak lensing analysis ,
objects are weighted by the inverse square of the uncertainty in the shear measurement ( e.g. , see hoekstra et al .
2000 , 2002a ) .
as more distant galaxies are fainter , they tend to have somewhat lower weights and the effective redshift distribution is changed slightly .
the dashed histogram in figure [ zdist]a shows the distribution weighted by the uncertainty in the shape measurement for each redshift bin .
the best fit parameterized redshift distribution has parameters @xmath38 , @xmath39 ( fixed ) and @xmath40 , which yields @xmath41 , only slightly lower than the unweighted case .
hoekstra et al .
( 2002b ) presented constraints on the matter density @xmath42 and the normalization of the power spectrum @xmath43 by comparing cold dark matter predictions to the observed lensing signal caused by large scale structure .
the derived value for @xmath43 depends critically on the adopted redshift distribution .
hoekstra et al .
( 2002b ) used galaxies with @xmath44 and a redshift distribution given by @xmath45 , @xmath46 and @xmath47 , which yields a mean redshift of @xmath48 .
these parameters were based on a comparison with redshift distributions determined from the hubble deep fields .
it is useful to examine how these assumptions compare to the rcs photometric redshift distribution for galaxies with @xmath44 , as displayed in figure [ zdist]b .
the best fit model , indicated by the smooth curve , has parameters @xmath49 , @xmath50 and @xmath51 , implying a mean redshift of 0.65 , about 10% higher than used by hoekstra et al .
( 2002b ) .
it is important to note , however , that the training set lacks a large number of objects beyond @xmath52 and @xmath53 . despite these shortcomings ,
the mean redshift of sources appears higher than what was used in hoekstra et al .
( 2002b ) , thus suggesting that their value for @xmath43 needs to be revised downwards .
the suggested change in source redshift could reduce the value for @xmath43 from hoekstra et al .
( 2002b ) by about @xmath54 to @xmath55 .
unfortunately it is not possible to robustly quantify the size of the revision .
we stress that without further work on photometric redshifts for faint , high redshift galaxies , it will be difficult to interpret current and , most importantly , future cosmic shear results .
the galaxy - galaxy lensing signal examined in this paper is much less sensitive to the uncertainty in the redshift distribution of faint , distant galaxies , as most of the signal is caused by lenses at much lower redshifts .
as mentioned above , to minimize uncertainties in our results further , we only use background galaxies with redshifts less than 1 , and select a sample of lenses with redshift @xmath3 .
hsieh et al . ( 2005 ) present various tests of the accuracy of the photometric redshifts .
comparison to the available spectroscopic data as well as comparing to other published distributions provides a clear way to quantify the uncertainties . in this section
we discuss some additional tests , based on the fact that the amplitude of the lensing signal is a well known function of the source redshift .
such a test provides a useful `` sanity '' check on the validity of the photometric redshift distribution .
the azimuthally averaged tangential shear @xmath56 as a function of distance from the lens is a useful measure of the lensing signal ( e.g. , miralda - escud 1991 ) : @xmath57 where @xmath58 is the mean surface density within an aperture of radius @xmath59 , and @xmath60 is the mean surface density on a circle of radius @xmath59 .
the convergence @xmath61 , or dimensionless surface density , is the ratio of the surface density and the critical surface density @xmath62 , which is given by @xmath63 where @xmath64 is the angular diameter to the lens . @xmath65 and @xmath66 are the angular diameter distances from the observer to the source and from the lens to the source , respectively .
it is convenient to define the parameter @xmath67,\ ] ] which is a measure of how the amplitude of the lensing signal depends on the redshifts of the source galaxies .
for instance , in the case of a singular isothermal sphere ( sis ) model , the dimensionless surface density is @xmath68 where @xmath69 is the einstein radius . under the assumption of isotropic orbits and spherical symmetry ,
the einstein radius ( in radians ) is related to the velocity dispersion and @xmath70 through @xmath71 to test the photometric redshifts from the rcs , we use galaxies with photometric redshifts @xmath3 to define a sample of lenses .
we compute the ensemble averaged tangential shear around these galaxies ( i.e. , the galaxy - mass cross - correlation function ) as a function of source redshift .
brighter galaxies are expected to be more massive , and should be given more weight . to derive the lensing signal , we assume that the velocity dispersion scales with luminosity as @xmath72 , a choice which is motivated by the oberved slope of the @xmath0-band tully - fisher relation ( e.g. , verheijen 2001 ) .
we select bins with a width of 0.1 in redshift , and measure the galaxy - mass cross - correlation function ( e.g. , see hoekstra et al .
2004 ; sheldon et al . 2004 ) out to 10 arcminutes .
this signal arises from the combination of the clustering properties of the lenses and the underlying dark matter distribution . in the remainder of the paper , while studying the properties of dark matter halos around galaxies , we limit the analysis to smaller radii and to ` isolated ' lenses .
however , by extending the range of measurements in this section , the signal - to - noise ratio is higher .
the signal is well described by a sis model for this range of scales ( as suggested by the reduced @xmath73 values for the fits ) .
the resulting value for the einstein radius as a function of redshift for the background galaxies is presented in figure [ re_zbg ] .
we find a negligible lensing signal for galaxies at the redshift of the lenses , whereas it increases for more distant sources .
for a given cosmology and a pair of lens and source redshifts the value of @xmath70 can be readily computed . however , the errors in the photometric redshift determination complicate such a simple comparison between the expected signal and the results presented in figure [ re_zbg ] . as was the case for the photometric redshift distribution
, the redshift errors will change the signal .
for instance at low redshifts , higher redshift galaxies will scatter into this bin , thus increasing the lensing signal . at higher redshifts , lower redshift object will scatter upwards , lowering the signal .
when comparing the observed signal to the signal expected based on the adopted @xmath74cdm cosmology we need account for these redshift errors . to this end
, we create simulated catalogs .
the first step is to compute a model lensing signal based on the observed photometric redshifts ( which are taken to be exact ) .
we then create a mock catalog by adding the random errors to the redshift ( while leaving the lensing signal unchanged ) .
these random errors are based on the observed distribution ( see e.g. , fig .
[ dzlens ] ) .
we measure the lensing signal as a function of redshift in the mock catalog .
this signal , indicated by the solid line in figure [ re_zbg ] can be compared directly to our actual measurements as it now includes the smoothing effect of redshift errors .
figure [ re_zbg ] shows that it traces the observed change in amplitude of the lensing signal very well .
another useful experiment is to measure the lensing signal when the lenses and sources are in the same redshift bin .
we note that this procedure enhances the probability that we measure the signal for galaxies which are physically associated .
if satellite galaxies tend to be aligned tangentially ( radially ) this would also lead to a positive ( negative ) signal .
the results from bernstein & norberg ( 2002 ) , based on an analysis employing spectroscopic redshifts , has shown that intrinsic tangential alignments are negligible .
unfortunately , in our case , the much larger photometric redshift errors ( compared to spectroscopic redshifts ) effectively suppress this potentially interesting signal .
in addition , for the rcs data , the interpretation of the signal requires a large set of spectroscopic redshifts to quantify the contributions of unassociated galaxies in each bin .
instead , we use these measurements as a test of the photometric redshifts .
the results are presented in figure [ re_zbin ] .
panel a shows the results for the tangential shear , whereas panel b shows the results when the background galaxies are rotated by 45@xmath75 ( which is a measure of systematics ) . in both cases
, we do not observe a significant signal .
the lack of a signal in the tangential shear in this case implies that the errors in photometric redshifts are relatively small .
if this were not the case , and higher redshift galaxies would contaminate the samples at lower redshifts , we would expect to observe a positive signal .
one of the major advantages of weak gravitational lensing over dynamical methods is that the lensing signal can be measured out to large projected distances from the lens .
however , at large radii , the contribution from a particular galaxy may be small compared to its surroundings : a simple interpretation of the measurements can only be made for ` isolated ' galaxies . in practice , galaxies are not isolated , which is particularly true for bright , early - type galaxies . in their analysis of sdss data , guzik & seljak ( 2002 )
quantified the contribution from clustered galaxies using a halo - model approach .
as discussed in 5 , we follow a different approach by selecting relatively isolated galaxies . as a result ,
our results are not strictly valid for the galaxy population as a whole .
nevertheless , the selection procedure is well defined and can be readily implemented when comparing to numerical simulations .
we limit the analysis to relatively small distances from the lens , thus ensuring that the signal is dominated by the lens itself . as a result
, we need to adopt a model for the mass distribution to relate the lensing signal to the mass of the lens .
our choice is motivated by the results of cold dark matter ( cdm ) simulations .
collisionless cold dark matter provides a good description for the observed structures in the universe .
numerical simulations , which provide a powerful way to study the formation of structure in the universe , indicate that on large scales cdm gives rise to a particular density profile ( e.g. , dubinski & carlberg 1991 ; navarro , frenk , & white 1995 , 1996 , 1997 ; moore et al .
we note , however , that there are still uncertainties regarding the slope at small radii and the best analytical description of the profile ( e.g. , moore et al .
1999 ; diemand et al .
2004 ; hayashi et al .
2004 ; tasitsiomi et al .
furthermore there is considerable scatter from halo to halo in the simulations .
our observations , however , can not distinguish between these various profiles , and instead we focus on the commonly used nfw profile , given by @xmath76 where @xmath77 is the virial mass , which is the mass enclosed within the virial radius @xmath78 .
the virial radius is related to the ` scale radius ' @xmath79 through the concentration @xmath80 . the function @xmath81 .
one can fit the nfw profile to the measurements with @xmath77 and concentration @xmath82 ( or equivalently @xmath79 ) as free parameters .
however , numerical simulations have shown that the average concentration depends on the halo mass and the redshift .
hoekstra et al .
( 2004 ) constrained the mass and scale radius of the nfw model using a maximum likelihood analysis of the galaxy - galaxy lensing signal , and found that the results agreed well with the predictions from simulations .
we therefore adopt the results from bullock et al .
( 2001 ) , who found from simulations that @xmath83 it is good to note that individual halos in the simulations have a lognormal dispersion of approximately 0.14 around the median . for the virial mass estimates presented here , we will use this relation between mass and concentration , thus assuming we can describe the galaxy mass distribution by a single parameter . by definition , the virial mass and radius
are related by @xmath84 where @xmath85 is the mean density at the cluster redshift and the virial overdensity @xmath86 , with @xmath87 ( bryan & norman 1998 ) . for the @xmath74cdm cosmology considered here , @xmath88 .
we also note that for the adopted @xmath74cdm cosmology the virial mass is different from the widely used @xmath89 .
this mass is commonly defined as the mass contained within the radius @xmath90 , where the mean mass density of the halo is equal to @xmath91 ( i.e. , setting @xmath92 and @xmath93 in eqn .
note , however , that other definitions for @xmath89 can be found in the literature as well .
the expressions for the tangential shear and surface density for the nfw profile have been derived by bartelmann ( 1996 ) and wright & brainerd ( 2000 ) and we refer the interested reader to these papers for the relevant equations .
as discussed above , we study a sample of lenses with photometric redshifts @xmath3 and @xmath24 .
this first selection yields @xmath94 lenses .
we split this sample in a number of luminosity and color bins and determine the virial radii from an nfw model fit to the observed lensing signal . for bright galaxies the lensing signal on small scales
is typically dominated by the dark matter halo associated with that galaxy . in the case of faint ( low mass ) galaxies , however , the signal can easily be dominated by contributions from a massive neighbor .
note that this neighbor need not be physically associated with the lens , as all matter along the line - of - sight contributes to the lensing signal .
we can study the relevance of the local ( projected ) density by measuring the lensing signal around a sample of ` faint ' lenses ( @xmath95 ) , as a function of the projected distance to the nearest ` bright ' lens ( @xmath96 ) .
this distance can be used as a crude measure of the density around the faint lens ( i.e. , the smaller the distance , the higher the density ) . to this end
, we split this sample of ` faint ' lenses into subsets based on their distance to the nearest bright galaxy .
we fit a sis model to the ensemble averaged lensing signal out to 2 ( @xmath97kpc at the mean distance of the lenses ) for each bin .
the reduced @xmath73 values for the best fit are all close to unity , indicating that the sis model provides a good fit to these observations .
we found that limiting the fit to smaller radii did not change the results apart from increasing the measurement errors .
figure [ re_dist ] shows the derived value for the einstein radius as a function of the distance to the nearest bright galaxy .
the results show a clear increase in lensing signal as the separation decreases , i.e. as the density increases .
however , at separations larger than @xmath98 , the observed lensing signal appears to be independent of the density .
larger data sets are required to make more definitive statements , but these findings suggest that we can measure the properties of ` isolated ' faint galaxies by limiting the sample to galaxies which are more than 30 arcseconds away from a brighter galaxy . although this is a rather strict selection for the faintest galaxies , bright galaxies can be surrounded by many faint galaxies and consequently are not truely isolated . in the remainder of this paper
, we present results based on the sample of ` isolated ' galaxies , unless specified otherwise .
this selection reduces the sample of lenses to 94,509 galaxies .
we split the sample of ` isolated ' lens galaxies into seven luminosity bins and measure the mean tangential distortion as a function of radius out to 2 arcminutes .
we fit an nfw profile to these measurements , with the virial mass as a free parameter , as described in 4 .
figure [ gtprof ] shows the measurements of the tangential shear as a function of projected distance from the lens for the seven @xmath2-band luminosity bins .
the results for the @xmath0 and @xmath1 filters are very similar to the ones presented in figure [ gtprof ] . to account for the fact that the lenses span a range in redshift , we have scaled the signal such that it corresponds to that of a lens at the mean lens redshift ( @xmath99 ) and a background galaxy at infinite redshift . in each panel in figure [ gtprof ] the average restframe @xmath2-band luminosity for each bin is indicated ( in units of @xmath100l@xmath101 ) .
the vertical scales in each of the panels in figure [ gtprof ] are the same , and as the luminosity of the lenses increases we observe a clear increase in the strength of the lensing signal .
the best fit nfw models for each bin are indicated by the dotted curves .
note that the current observations can not distinguish between an nfw profile ( used here ) and other profiles such as the sis model .
there are more faint galaxies relative to the number of bright galaxies and the errors in the photometric redshift estimates will have as the net effect of faint galaxies getting scattered to higher luminosity bins , hence biasing the mass at fixed luminosity to a lower value . to estimate the level of this bias we create mock catalogs .
we assume a power - law mass - luminosity relation and compute the model lensing signal using the observed photometric redshifts of the lens and source galaxies .
we analyse this ` perfect ' catalog and compute the virial masses as a function of luminosity .
we then use the observed photometric redshift error distribution as a function of apparent magnitude ( see figure [ dzlens ] ) to create a number of new catalogs where the random error is added to the redshift ( note that the lensing signal is not changed ) .
these catalogs are also analysed and yield the ` observed ' virial mass as a function of luminosity .
as expected , the resulting masses are smaller than the input masses and the change in mass depends on the luminosity .
the results are presented in figure [ corfac ] for the @xmath0 , @xmath1 , and @xmath2-band .
different choices for the mass - luminosity relation ( within reasonable bounds ) yield very similar curves . to infer the correct virial mass , we scale the observed virial masses by these curves . [ cols="^,^,^,^,^,^ " , ] observations of the cosmic microwave background ( e.g. , spergel et al .
2003 ) have yielded accurate measurements of the baryon fraction in the universe . based on wmap observations , spergel et al .
( 2003 ) obtained @xmath102 and @xmath103 .
if we assume that baryons do not escape the dark matter overdensity they are associated with , the ratio of mass in baryons to the total mass of the halo is @xmath104 . in the following , we will also assume @xmath105=71/km / s / mpc , which is the currently favoured value .
for the pegase salpeter model , this implies that the fraction of the mass in stars is @xmath106 ( average of the @xmath0 and @xmath2 value ) , whereas the scaled salpeter imf yields a lower value of @xmath107 .
comparison with the value of @xmath108 from cmb measurements suggests that only @xmath109 and @xmath110 of the baryons are converted into stars for the standard and scaled salpeter imfs respectively .
the actual results for the two filters considered here are indicated separately in table [ tab_star ] by @xmath111 .
table [ tab_star ] also lists the average results when we consider blue and red galaxies separately .
the implied star formation efficiencies for early - type galaxies are low .
we note that similar efficiencies have been inferred for galaxy clusters ( e.g. , lin , mohr & stanford 2003 ) .
interestingly , our results imply that late - type galaxies convert a @xmath112 times larger fraction of baryons into stars .
this result is robust , as it does not depend much on the adopted imf .
guzik & seljak ( 2002 ) also found a factor of @xmath112 difference in star formation efficiency between early and late - type galaxies , in good agreement with our findings .
hence , these results provide very important , direct observational constraints on the relative star formation efficiency during galaxy formation for different galaxy types .
these findings suggest that the mechanism for the formation of early type galaxies is somehow more efficient in removing gas compared to late type galaxies .
ram pressure stripping might be more prevalent , given that early type galaxies are typically found in high density regions , or they might form while developing strong winds that blow out most of the baryons . irrespective of the process responsible for ejecting baryons , the resulting galaxy will always have a stellar fraction which is greater or equal to the fraction of stars in its progenitors .
if we consider the situation where early - type galaxies form through mergers , it is clear that not all early - type galaxies can be the result of merging the late - type galaxies studied in this paper .
ejecting @xmath113 of the stars during the merger process might seem an option , but this is hard to envision without removing a similar fraction of the dark matter halo .
hence , the progenitors of early - type galaxies must have had a low fraction of their mass in stars .
this could be achieved if early - type galaxies ( or their progenitors ) formed early on without forming new stars at later times ( because they lost their gas ) and if later type galaxies sustained their star formation for a much longer time , thus building up a larger fraction of mass in stars .
recent estimates of the star formation rates of high redshift galaxies , suggest a qualitatively similar picture , in which early - type galaxies formed the bulk of their stars very early on with a sharp drop in star formation rates at @xmath114 , and less massive ( late - type ) galaxies continue to form most of their stars at a later time and over a much longer period of time ( e.g. , mccarthy et al . 2004 ; juneau et al . 2005 ) .
in addition to stars , galaxies contain some gas , which needs to be included if we are to do a full accounting of the visible baryon contents of galaxies .
although the amount of molecular hydrogen is uncertain , the amount of neutral hydrogen is relatively well determined from 21 cm line studies .
the relative amount of hi gas is a function of galaxy type , with late - type galaxies being more gas rich .
we use the results from roberts & haynes ( 1994 ) for the @xmath115 ratio to estimate the amount of mass in gas ( correcting the hydrogen mass to account for the primoridial helium abundance ) .
the inclusion of gas slightly raises the mass in detected baryons for the bluest galaxies but this component is negligible for the red galaxies . adding estimates for the amount of molecular hydrogen does not change the numbers either .
the resulting fraction of the mass in baryons in galaxies , @xmath116 , is listed in table [ tab_star ] as well . only for the blue galaxies , under the assumption of a standard salpeter ( 1955 ) imf ,
is the baryon fraction marginally consistent with the value determined from observations of the cmb ( spergel et al .
however , as discussed earlier , the results for this model should be considered upper limits to the baryon fraction , given that the stellar mass - to - light ratios are too high to fit rotation curves ( e.g. , bell & de jong 2001 ) .
the results from the scaled imf from bell & de jong are probably more representative of the actual baryon fractions in galaxies , thus implying that a significant fraction of the gas must have been lost .
we have measured the weak lensing signal as a function of restframe @xmath0 , @xmath1 , and @xmath2-band luminosity for a sample of ` isolated ' galaxies , with photometric redshifts @xmath3 .
this selection of relatively isolated galaxies minimizes the contribution of group / cluster halos and nearby bright galaxies .
the photometric redshifts were derived by hsieh et al .
( 2005 ) using @xmath117 photometry from the red - sequence cluster survey . to add to the extensive study described in hsieh et al .
( 2005 ) , we confronted the photometric redshifts to tests that are unique to weak lensing .
these results showed that the lensing signal around a sample of foreground galaxies scales with source redshift as expected .
the photometric redshift distribution determined by hsieh et al .
( 2005 ) suggests that the mean redshift of galaxies used in the measurement of the lensing signal by large scale structure ( hoekstra et al . 2002a ; 2002b )
is somewhat higher than previously assumed .
if correct , this would imply a somewhat lower value for the normalization of the matter power spectrum , @xmath43 , compared to the published results .
the difference is expected to be less than @xmath118 , but with the current data we can not reliably quantify the size of the change .
virial masses were determined by fitting an nfw model to the tangential shear profile .
note , that intrinsic scatter in the mass - luminosity relation will result in an underestimate of the mean virial mass for a galaxy of a given luminosity , as suggested by tasitsiomi et al .
( 2004b ) .
the magnitude of this effect depends on the assumed scatter , and we have ignored this in our analysis .
we found that the virial mass as a function of luminosity is well described by a power - law with a slope of @xmath119 , with similar slopes for the three filters considered here .
this result agrees with other observational studies ( guzik & seljak , 2002 ) and predictions from semi - analytic models of galaxy formation ( e.g. , kauffmann et al . 1999 ; benson et al .
2000 ; van den bosch et al .
2003 ) . for a galaxy with a fiducial luminosity of @xmath4l@xmath5 we obtained a mass @xmath120m@xmath7 . converting this result to match the filter and definition for the mass used by guzik & seljak ( 2002 ) , yields a mass of @xmath121 for a galaxy with a luminosity of @xmath122 , in agreement with guzik & seljak ( 2002 ) , who found @xmath123 .
we examined the efficiency with which baryons are converted into stars . to do so
, we used the restframe @xmath124 color as a measure of the mean stellar mass - to - light ratio .
the color also provides a crude indicator of galaxy type ( e.g. , robert & haynes 1994 ) .
we considered a standard and a scaled salpeter imf ( see bell & de jong 2001 ) .
the latter is more realistic , whereas the former yields stellar mass - to - light ratios that are too high to fit rotation curves of spiral galaxies .
irrespective of the adopted imf , we found that the stellar mass fraction is about a factor of two lower for early - type galaxies , as compared to late - type galaxies . including the fraction of baryons in gas only increases the fraction of observed baryons slightly .
hence , our results suggest that galaxy formation is very inefficient in turning baryons into stars and in retaining baryons .
these results provide important , direct observational constraints for models of galaxy formation . under the assumption that the scaled salpeter imf is correct ,
our results imply that late - type galaxies convert @xmath9% of baryons into stars .
early - type galaxies do much worse , with an efficiency of @xmath125% .
this implies that the progenitors of early - type galaxies have a low fraction of their mass in stars .
a possible explanation of this result is that early - type galaxies formed early on and stopped forming new stars , because they lost most of their baryons ( e.g. , through winds or ram pressure stripping ) .
if later type galaxies , on the other hand , continued to form stars this would lead to a higher stellar mass fraction .
such a scenario is , at least qualitatively , in agreement with recent estimates of the star formation rates of high redshift galaxies ( e.g. , mccarthy et al .
2004 ; juneau et al . 2005 ) . | we have measured the weak lensing signal as a function of restframe @xmath0 , @xmath1 , and @xmath2-band luminosity for a sample of ` isolated ' galaxies .
these results are based on four - band photometry from the red - sequence cluster survey , enabling us to determine photometric redshifts for a large number of galaxies .
we select a secure sample of lenses with photometric redshifts @xmath3 and study the relation between the virial mass and baryonic contents .
in addition , we discuss the implications of the derived photometric redshift distribution for published cosmic shear studies . the virial masses are derived from a fit to the observed lensing signal . for a galaxy with a fiducial luminosity of @xmath4l@xmath5
we obtain a mass @xmath6m@xmath7 .
the virial mass as a function of luminosity is consistent with a power - law @xmath8 , with similar slopes for the three filters considered here .
these findings are in excellent agreement with results from the sloan digital sky survey and semi - analytic models of galaxy formation .
we measure the fraction of mass in stars and the baryon fraction in galaxies by comparing the virial mass - to - light ratio to predicted stellar mass - to - light ratios .
we find that star formation is inefficient in converting baryons into stars , with late - type galaxies converting @xmath9% and early - type galaxies converting only @xmath10% of baryons into stars .
our results imply that the progenitors of early - type galaxies must have low stellar mass fractions , suggestive of a high formation redshift . |
compact groups of galaxies are found at the extreme end of the distribution of galaxy surface densities ; by definition , compact groups consist of four or more isolated galaxies found within a small area on the sky . using the @xcite compact group selection criteria , the surface densities of galaxies in compact groups are similar to or larger than those found in the centers of massive galaxy clusters . a radial velocity survey of the 100 hickson compact groups ( hcgs ; * ?
* ) has verified that the majority of these hcgs are physical associations of 3 or more galaxies with a median velocity dispersion of 200 km sec@xmath0 .
one expects that in such dense groups with small velocity dispersions that interactions and mergers among the members are inevitable .
as expected , the morphologies of the galaxies in compact groups often show evidence of tidal interaction .
galaxy mergers are complicated phenomena , and it is difficult to disentangle the past histories of the merging galaxies , even in isolated , merging pairs . in compact groups
there is often evidence for multiple interactions ( e.g. , * ? ? ?
* ) , so unravelling the history of these systems is even more of a challenge .
fortunately , star formation is a useful tool for dating some of the discrete events in the interaction history of a galaxy merger .
relying on _ hubble space telescope ( hst ) _
images , a growing number of studies have found evidence for compact star cluster formation in systems of interacting galaxies or merger remnants @xcite . using stellar population synthesis models , the ages of the observed young star clusters can be estimated from their photometric colors . in cases where discrete populations of young clusters are found in different regions of an interaction or merger remnant , variations among the ages of the clusters
can be used to date events in the merger history .
for example , @xcite identified four distinct populations of young star clusters in the `` antennae '' ( ngc 4039/39 ) , and used the ages of these populations to infer some of the evolutionary history of this system . @xcite
have identified a number of compact clusters in hcg 92 , stephan s quintet .
this system is more complicated than the antennae , but the star cluster ages have been used to identify distinct epochs of cluster formation and to date some of the interaction events .
star formation in interacting galaxies is not necessarily limited to compact cluster formation ; there is also observational evidence for the formation of extended , dwarf galaxy - sized objects in the tidal debris of a number of systems ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
@xcite have performed a statistical analysis of a large sample of hcgs and found evidence that tidal dwarf galaxy formation in compact groups may be common .
numerical simulations suggest that dwarf galaxy formation in tidal tails is possible @xcite ; however , it is unclear whether the observationally identified `` tidal dwarf galaxies '' ( tdgs ) will evolve into the bound entities seen in the simulations .
recently , a few groups have presented dynamical analyses of several tdgs @xcite in order to assess the likelihood that these objects will remain bound , but the results remain ambiguous .
_ hst _ imaging of hcgs is useful for addressing both the interaction history of a compact group as well as the presence and nature of any putative tdgs . in this paper
, we present wide field planetary camera 2 ( wfpc2 ) observations of hickson compact group 79 , seyfert s sextet , taken in the @xmath1 , @xmath2 , @xmath3 , and @xmath4 ( f336w , f439w , f555w , and f814w ) filters . in sections 2.22.4 ,
the compact clusters in the group are identified and their ages estimated from stellar population models . in sections 2.52.6 ,
we present the extended objects in the field and discuss their nature .
section 3 discusses a spectroscopic follow - up observation of one candidate dwarf galaxy associated with the sextet . finally , in section 4
, we compare the population of compact clusters and dwarf galaxies in seyfert s sextet to those found in the recent _ hst _ studies of hcg 92 ( stephan s quintet ; * ? ? ? * ) and hcg 31 @xcite .
@xcite was the first to identify and study this `` unusually densely crowded group of six galaxies '' from photographic plates taken with the harvard schmidt telescope .
while the second article refers to six galaxies and does name the group a sextet , @xcite mentions that baade s observations of the group show that the object labeled `` ngc6027e '' . ]
is `` not a separate galaxy but a tidally distorted part of ngc6027 . ''
@xcite goes on to say that `` if this is actually the case , we have a very extraordinary filament which not only has a perceptible condensation but a filament which is nearly twice as large as the parent spiral . ''
based on the morphological differences between the two late - type spirals and the tidally distorted early - type spirals , @xcite considered two possibilities : ( 1 ) the sextet is truly comprised of three physically associated early - type galaxies , two background spirals , and one bright tidal tail , or ( 2 ) it is a physically associated group of six galaxies .
seyfert s sextet fits the criteria for inclusion in the hickson compact group catalogue @xcite .
it is the most compact of the hcgs @xcite , with a median galaxy separation of @xmath5 kpc ( we adopt @xmath6 km s@xmath0 mpc@xmath0 throughout this paper ) . the group is now known to consist of four galaxies with accordant redshifts , one bright tidal tail , and a fifth galaxy at a redshift much larger than the group median redshift @xcite . using the mean redshift of the accordant galaxies ,
@xmath7 @xcite , the distance to the sextet is 57.5@xmath8 mpc and the distance modulus is 33.8 magnitudes .
a summary of the global properties of seyfert s sextet is given in table [ ssparams ] . in figure
[ dssimage ] , we present an image of seyfert s sextet from the second generation red digitized sky survey that includes labels for the individual member galaxies using the scheme of @xcite . the wfpc2 instrument on the _ hubble space telescope _
was used to observe seyfert s sextet in 2000 june ( program i d 8717 , hunsberger pi ) .
the observations were planned in such a way as to place the sextet galaxies on the three wide field chips . because of the placement of the galaxies of interest on the three wf chips , and since additional read noise in the pc data raises the detection threshold for point sources on this chip compared to the wf chips , the planetary camera data were not included in this study .
the total exposure time in each filter was broken up into four cosmic ray - split exposures .
one dithering offset of @xmath92.5 pixels in the @xmath10 and @xmath11 directions was introduced so that two of the four subexposures are at one dither location and the other two subexposures are at another .
the exposure times were @xmath12 seconds in f814w and f555w and @xmath13 seconds in f439w and f336w .
the gain was set to 7 @xmath14/adu during these exposures . in order to use the latest _
calibration , the data were recalibrated with the `` best '' reference files available from the _ hst _ archive in 2002 january . known
hot pixels were removed from the images using the iraf stsdas task warmpix .
after hot pixel removal , the two images at each dither position in each filter were averaged together with gcombine and then further cleaned of cosmic rays with the iraf task cosmicrays . identification and photometry of point sources ( see 2.2 ) were done individually on each of the images of the two dither locations and the results subsequently averaged .
the choice to analyze each dither location separately was made because the detection of extremely faint sources and the study of the sizes of these objects are not the focus of this paper . using the iraf tasks geomap and geotran to solve for the transformation between the two dither locations , a final , combined image in each filter that retains the original pixel scale
was created .
identification and photometry of the extended sources in the field were done on the final , combined image ( see 2.5 ) .
also , a color image of seyfert s sextet ( figure [ primage ] ) was created by combining the black and white data collected through the 4 filters . the intensities were first clipped and scaled to bring out faint emission .
( karma s kview task was employed for this step but the remainder of the manipulations were accomplished using gimp . ) each filter s scaled image was then assigned a color that approximated the peak wavelength of that filter : f336w was colored reddish - violet , f439w blue , f555w yellowish - green , and f814w red .
these colour images were combined using the `` screen '' algorithm in gimp .
the contrast of the combined image was adjusted to produce a neutral , dark background and emphasize particular features .
it is the inclusion of the ultraviolet filter that causes the disk of the central spiral galaxy , as well as the central regions of the edge - on spiral , to be rendered more pink than they would be if strictly optical filters were used .
the method used for point source detection follows very closely the technique used by @xcite .
a brief summary of this method is presented here : each dither location in each filter was initially searched for point sources using the daofind task within iraf s implementation of daophot @xcite .
the detection threshold was set very low in order to find potential sources in regions of higher than average background ( e.g. , in the faint tidal tail ) . in order to remove some spurious detections , only those objects detected at the same location in both dither images
were retained .
initial aperture photometry was performed on this large number of objects ( thousands per wf chip ) , and all objects with @xmath15 at either or both dither locations were thrown out .
although this process serves to eliminate most of the spurious sources , a number of objects remain in the catalogue that can be eliminated as potential compact star clusters associated with seyfert s sextet . adopting the mean group redshift as a distance measure , the sextet is @xmath958 @xmath16 mpc from earth , and at this distance , one pixel is @xmath927 @xmath16 pc .
typical galactic globular clusters have @xmath17pc @xcite , and even large , young clusters seen in the antennae have effective radii ( the radius enclosing half of the light , seen in projection ) of @xmath18 pc , which leads us to expect that compact clusters in the sextet will be unresolved .
thus , we can clean the catalogue further by eliminating all resolved sources .
resolved sources were removed from the catalogue using the same set of tests used by @xcite ( which are derived from those in * ? ? ?
specifically , at each dither location , every potential point source was photometered with an aperture radius of 0.5 pixels and a second time with an aperture radius of 3.0 pixels .
the difference in the @xmath3 magnitude between the two apertures , @xmath19 was calculated , and all objects with @xmath20 ( likely cosmic rays or other hot pixels ) or @xmath21 ( resolved sources ) in either pointing were flagged . also , using the iraf task radprof , fifth order cubic splines were fit to the radial profiles of each potential point source . in all four filters ,
the mean fwhm for the likely point sources is @xmath22 pixels , however the peak of the four distributions are near 0.9 pixels , with a non - gaussian tail to higher values . to separate real point sources from resolved sources
, we chose to flag those objects with fwhm @xmath23 pixels , also .
each object was thus tested four times : once with the @xmath19 test and once with the fwhm test at each dither location .
any object with two or more flags was removed from the catalogue of point sources .
the final catalogues did contain some sources that passed all algorithmic tests , but could still be easily ruled out as potential compact star clusters .
these included bright , foreground , galactic stars , the unresolved nuclei of a few background galaxies , and regions in the disk of the background galaxy in the sextet ( ngc 6027e ) .
the foreground stars and galactic nuclei are easily distinguishable from the candidate star clusters .
although the regions in ngc 6027e are blue point sources that might be mistaken for star cluster candidates , they are all aligned with the prominent spiral arms in this disk galaxy , so we are confident that they are unlikely to be point sources associated with seyfert s sextet .
we removed by hand these three types of contaminating objects from the catalogue . in the end , we were left with a total of 188 star cluster candidates ( see figure [ fchart1 ] for a finding chart ) detected in at least one filter .
very few objects were detected in the two blue filters , f439w and f336w ; we discuss this further in 2.4 .
the exposures of seyfert s sextet were taken at two dither locations and can be combined in order to enhance the angular resolution using the `` drizzle '' software developed by @xcite . since the cluster population of seyfert s sextet should be unresolved even in drizzled images , we chose to apply our cluster detection algorithm to the raw , undrizzled images rather than to drizzled images .
however , we did combine the f439w and f555w images of the wf2 chip with the drizzle software in order to test the differences in cluster detection on drizzled and undrizzled images .
drizzle was run with the parameters @xmath24 and @xmath25 , which is equivalent to the `` shift and add '' technique , and the resulting images have a pixel scale twice as fine as that of the input images .
we used an identical cluster detection algorithm to the one described above : the objects detected by daofind in the f439w and f555w drizzled and undrizzled images were pruned first by signal - to - noise and then were pruned further using the fwhm and @xmath26 ( the aperture sizes were doubled to 1.0 pix and 6.0 pix for the drizzled image ) tests discussed in detail in the previous section .
the most significant cut is the signal to noise cut ; since the threshold used in daofind is so low , many of the objects detected have @xmath27 .
furthermore , there are significant ( at the @xmath28% level ) differences in @xmath29 between the objects in the drizzled image and in the undrizzled image .
this is especially true of objects found in regions where the background is variable , such as within the bright isophotes of ngc6027a and ngc6027d . after applying our signal to noise criterion ,
a number of ( likely spurious ) objects within the bright isophotes of these two galaxies remain in the catalogue created from the undrizzled image .
however , in comparison , many of these get removed from the catalogue created from the drizzled image .
the second criterion used to flag objects for removal from our point source catalogues are the `` structure '' parameters fwhm and the concentration or @xmath26 .
what we find is that objects that have @xmath30 lie in a narrow range of fwhm , indicating that they are likely to be point sources .
almost every object within this window of concentration parameter have fwhm @xmath31 pixels in the undrizzled image and fwhm @xmath32 pixels in the drizzled image . for objects with @xmath33 , the range of fwhm found for objects in both the drizzled and undrizzled images is much larger , extending from small values typical of point sources all the way to fwhm @xmath915 pixels or so .
thus , it appears that @xmath34 signals the transition from point sources to resolved sources .
our detailed comparison between drizzled and undrizzled images of the wf2 chip in filters f439w and f555w shows that in the end the point sources identified are nearly identical .
the largest difference found is in @xmath29 , and results in more spurious objects being identified in the disk of ngc6027d and near the dust lane in ngc6027a in the undrizzled image than in the drizzled image .
although the pixel scale is different by a factor of two between the drizzled and undrizzled images , the results for the structure parameter are nearly identical for point source candidates in both images : those objects with @xmath35 have fwhm values expected for point sources in both the drizzled and undrizzled images .
those objects with @xmath36 have fwhm values typical of resolved objects .
thus , we conclude that for the dataset presented here , our point source detection algorithm is not significantly improved by the application of the drizzling technique .
we note that a recent study of the star cluster candidates in ngc 3610 by @xcite supports this conclusion . in a previous study of ngc 3610 , @xcite used undithered images to identify point sources using an algorithm almost identical to the one used here . in the more recent study
, @xcite drizzled together deeper exposures of ngc 3610 .
even though the @xmath29 and pixel scale were improved in the more recent images , the results of the new search were similar to the results of the original study .
they found that 22 of 23 cluster candidates identified from the earlier images were recovered in the new images .
also , 11 new , previously unidentified clusters were discovered in the higher resolution images .
however , the majority of these are within the disk of ngc 3610 , and the authors attribute their detection to their ability to identify clusters in regions of higher background , which is likely because of the combination of improved @xmath29 and pixel scale , not just because of the improved pixel scale .
we performed circular aperture photometry of all point sources detected using the method described previously .
for consistency with @xcite , the photometry of each source was calculated using similar techniques .
the technique used and the few differences between this study and @xcite are described here .
each object was measured with the iraf apphot package using a circular aperture of 2 pixel radius . the sky level for each object
was determined by taking the median value in an annulus of 7 pixel inner radius and 10 pixel outer radius centered on the object . a correction for charge transfer efficiency ( cte )
was calculated for each object using the formulation given in @xcite .
photometry was measured at both dither positions for all point sources , the cte correction was applied , and then the results for the two positions were averaged . in order to transform the instrumental magnitudes to the standard vegamag system , the zero points from @xcite were adopted .
these zero points were calculated using the standard @xmath37 aperture , and therefore , aperture corrections need to be applied to our data to account for the different aperture size used .
aperture corrections were calculated by photometering bright , isolated point sources in our 2 pixel aperture and in the 5.15 pixel , @xmath37 aperture .
the difference in magnitudes in these two measurements was calculated for each point source , and the mean adopted as the aperture correction .
there were not enough point sources to solve for the aperture correction for each wf chip independently , so we adopt the same aperture correction for each chip .
table [ apcor ] lists the adopted values for the aperture correction in each filter , the number of objects used to derive the correction , and the standard deviation of the measurements .
the value for the galactic extinction towards seyfert s sextet was derived from the @xcite dust maps .
the value returned was @xmath38 , which is adopted as the foreground reddening towards all objects in the wfpc2 images of the sextet . in order to determine the reddening corrections for the specific _ hst _ filters employed in this study , a synthesized instantaneous - burst star cluster model spectrum ( the bc95 models ; * ? ? ?
* ; * ? ? ?
* ) with an age of 10 gyr ( see 2.4 ) was reddened using a @xcite extinction law . using the stsdas synphot package ,
this reddened spectrum was convolved with the _ hst _ filter functions and the following corrections were calculated : @xmath39 , @xmath40 , @xmath41 , and @xmath42 .
all of the point source photometry presented here has been corrected using these estimates for the galactic reddening corrections .
figure [ photerr ] presents the calibrated photometry and associated measurement error for the 188 point sources identified in the three wide field chips . among the 188 point sources detected , 106
were simultaneously detected in both @xmath3 and @xmath4 , 26 in @xmath2 , @xmath3 , and @xmath4 , and only 5 were detected in all four filters .
figures [ cmd ] and [ ccd ] are the @xmath43 vs. @xmath3 color magnitude diagram ( cmd ) for the 106 sources detected in both filters and the @xmath44 vs. @xmath43 two color diagram ( 2cd ) for the 26 sources detected in all three filters .
although a correction for galactic reddening was applied to the point source photometry , differential reddening due to the dust content of the compact group itself is also likely to affect the photometry of the point sources . for reference , a reddening vector equivalent to one magnitude of extinction in the f555w filter
is plotted in figures [ cmd ] and [ ccd ] .
differential reddening is likely to be affecting many of the sources to some degree .
however , because at least some of the point sources are found far from the outer isophotes of the giant galaxies , it is unlikely that there is a large , systematic shift in the colors and magnitudes of all of the sources .
if we assume that the differential reddening in the group is not severe , then one conclusion that can immediately be drawn from the cmd and the 2cd is that there are very few bright , blue point sources detected .
our choice of exposure times for the two blue filters was guided by observations of blue point sources in other studies of mergers ( e.g. , the antennae , * ? ? ?
* ) and compact groups ( e.g. , hcg92 , * ? ? ?
given our exposure times , the resulting limiting magnitude in @xmath2 is brighter than our @xmath3 limit ( the magnitudes where the mean photometric error for point sources reaches 0.1 mag are 24.6 , 25.0 , and 24.3 for @xmath2 , @xmath3 , and @xmath4 , respectively ) , however , the limits are such that we should have detected any objects with @xmath45 and @xmath46 in both filters
. this observational selection effect does render us insensitive in @xmath2 to faint point sources with @xmath47 , however , the candidate young star clusters identified in other systems are all blueward of this limit . in seyfert s sextet , we find only a handful of objects are detected blueward of @xmath48 .
the majority of the point sources detected have @xmath49 ( which are likely to have @xmath50 ) , but most went undetected in @xmath2 .
this lack of bright , blue point sources in the sextet is somewhat unexpected since these objects are found in abundance in similar systems .
for example , in @xcite , the peak in the color histograms for the identified point sources in stephan s quintet ( hcg92 ) are found at @xmath51 and @xmath52 .
also , @xcite find a number of bright point sources in _ hst _ images of hcg31 that have @xmath53 ( the extremely blue objects in this sample are likely to have a large contribution to their flux from nebular line emission ) .
the 2cd for seyfert s sextet point sources ( figure [ ccd ] ) also includes the evolutionary tracks derived from the bc95 @xcite solar metallicity instantaneous burst stellar population models . in the cmd , the colors and magnitudes of these same bc95 models
are included , where the magnitudes have been normalized for a cluster with masses of @xmath54 and @xmath55 .
there is a significant amount of scatter in the data for the star cluster candidates , however , the majority of the objects detected have colors in the 2cd similar to the expectations for clusters of 1 gyr or older . in the cmd , the track for the @xmath55 clusters fits the data better than that of the lower mass model , and also indicates an age for the clusters in excess of 1 gyr .
unlike the two previous studies of star cluster formation in compact groups @xcite , there does not appear to be any significant population of candidate young star clusters with ages less than 1 gyr . in figure
[ hists ] , we present histograms of the @xmath56 colors for the 106 point sources detected in both @xmath3 and @xmath4 ( top panel ) and the @xmath3 magnitudes for the 84 likely cluster candidates in the restricted color range @xmath57 ( bottom panel ) .
the color and magnitude distributions presented in this figure are consistent with our interpretation of an older population of globular clusters .
the mean color of the objects in the sextet with @xmath58 is 0.99 , which is similar to the mean colors of the globular cluster populations in _ hst _ observations of ellipticals and s0s found by @xcite .
the histogram of the @xmath3 magnitudes of the likely cluster candidates identified in the sextet is consistent with the luminosity function of the globular cluster populations seen in the deeper photometry of @xcite ; some of the single elliptical galaxies in that study contain more clusters brighter than @xmath59 than are found in the entire compact group studied here .
however , since the observations presented in this paper were not designed to study the fainter old population of globular clusters in this compact group , it is likely that our census of the old clusters in the group is incomplete . given the likely incompleteness , though , the data on the sextet cluster candidates presented in figure [ hists ] are consistent with the standard ( old ) globular cluster luminosity function @xcite , not just the _ hst _ data of @xcite .
blue globular clusters are expected to be as bright or brighter than the older , red globular clusters ( unless they are of much lower mass ) ; for example , many of the blue clusters identified in ngc3256 @xcite have @xmath60 , which is above our detection threshold .
thus , figures [ cmd ] and [ hists ] should represent accurately the population of clusters in the sextet ; the population is truly dominated by red clusters .
however , among the sources that were detected in @xmath2 , @xmath3 , and @xmath4 , there are a few blue point sources observed , those on wf2 that have @xmath61 .
four of the five blue point sources are located within the disk of ngc6027d , and the fifth is just south of the disk ( it is the southernmost point source visible in figure [ fchart1 ] ) .
this region appears to be the only site that includes young , star - forming regions .
@xcite have identified samples of dwarf galaxy candidates in a number of hcgs using the focas software package @xcite to perform star / galaxy separation on ground - based @xmath62 band images .
seyfert s sextet was not included in this study , however , we present here ( figure [ kpnoimage ] ) a previously unpublished , deep @xmath62 band image of seyfert s sextet observed after the conclusion of the hcg survey of @xcite . although the majority of the hcgs in the @xcite sample were observed with the palomar 1.5 m , this image was taken with the kitt peak 0.9 m telescope in 1996 may .
the data reduction and analysis techniques used to process this image were identical to those used in @xcite . in figure [ kpnoimage ]
, a large circle indicates the boundary of the group ; the radius of this circle is twice the size of the radius of the smallest circle that encloses the 23 mag arcsec@xmath63 isophotes of the giant galaxy members , @xmath64 ( see * ? ? ?
* ) , and within that boundary every non - stellar object is indicated by an ellipse . apart from the objects that make up the sextet ( five galaxies and the bright tidal tail ) , 33 objects were detected and classified as non - stellar .
the majority of these objects are found in the northwest quadrant of the image , and several appear to be superposed on the fainter tidal tail , which extends to the northwest of ngc6027c .
comparison to the _ hst _ image ( figure [ primage ] ) reveals that the majority of the galaxies identified in the ground - based image of the sextet are background galaxies .
viewed at higher angular resolution , many of the non - stellar objects appear to be background disk galaxies , and the two bright , red galaxies west of ngc6027c appear to be background giant ellipticals .
however , several of the candidate dwarf galaxies do appear to be just that , dwarf galaxies associated with the sextet
. furthermore , several faint , diffuse objects not visible in the ground - based image are also apparent in the _ hst _ image . in this section
we will describe the identification and photometry of the extended sources observed in the _ hst _ image .
detection of candidate dwarf galaxies associated with seyfert s sextet is reasonably straightforward ; assuming a distance modulus of 33.8 , a typical dwarf galaxy will have @xmath65 , or a total magnitude of @xmath920 in the f555w filter .
furthermore , a dwarf galaxy with a core radius of @xmath9300 pc will have an angular radius @xmath66 , and will therefore cover roughly 300 or so pixels .
thus , we expect most dwarf galaxies ( including those fainter than @xmath65 ) to be detected with high signal - to - noise in our f555w and f814w images . there are some limits , though ; the photometric error is @xmath67 for point sources with @xmath68 , and the sensitivity of the wf chips to low surface brightness sources is limited .
thus , we do not expect to detect the faintest , diffuse dwarf galaxies , similar to the local group dwarf spheroidal galaxies ( which are as faint as @xmath69 ) .
selection of extended sources to photometer was accomplished using the focas software package , however , we set the parameters in order to select relatively bright , extended sources .
the catalogue was constructed to include all objects with a minimum area of 50 pixels ( 0.5 square arcseconds ) with fluxes 2.5@xmath70 above the background .
this catalogue did contain some obvious contaminants , including saturated stars and several resolved regions in ngc6027d and ngc6027e .
these were removed by hand .
ultimately , we were left with 38 objects .
most of the objects identified in the ground - based image that are within the boundaries of the wfpc2 image were included , as well as a number of fainter objects not visible in figure [ kpnoimage ] .
the only objects identified in the ground - based image that were not recovered by the automated detection algorithm were those found very near the boundaries of the wf chips , where the images had been trimmed .
we also note than in a few cases , objects identified from the ground as a single extended source were resolved into two extended sources in the _ hst _ images .
it is apparent by visual inspection that a number of the extended sources seen in figure [ primage ] are disk galaxies .
we remove from further consideration as potential dwarf members of seyfert s sextet those galaxies with clear spiral structure and those that appear to be edge - on spiral galaxies .
we assume that these objects are background galaxies .
for the final catalogue of candidate dwarf galaxies , we have photometered 29 objects ; six from the wf2 chip , 17 from wf3 , and six from wf4 .
in addition , we have photometered six of the background spiral galaxies ( the other three were found on the edges of two wf chips , in the region trimmed from the images during data reduction ) for comparison to the sample of dwarf galaxy candidates .
figure [ fchart2 ] is a finding chart for the 29 dwarf galaxy candidates and six background spiral galaxies that were photometered ; a contour plot of a smoothed version of figure [ primage ] ( smoothed to reduce some of the noise in the fainter isophotes ) is shown with square boxes marking each object in this sample .
for each extended object in the catalogue , the stsdas task ellipse was used to fit elliptical isophotes to the galaxy in the _ hst _ @xmath4 image out to a semi - major axis where the signal to noise became too low for the fitting routine to work successfully .
the ellipse parameters derived for each object from the @xmath4 image were then used to calculate the object s flux in the other filters .
a value of the average background flux per pixel was calculated for each image . for each object , the total flux within the elliptical apertures was corrected by subtracting off the total background contribution within the aperture , assuming that this average background flux per pixel is flat across the image .
after background subtraction , the total flux for each object was converted into a total magnitude using the adopted @xcite zero points ( which were corrected to infinite aperture using the standard 0.1 magnitude offset ; * ? ? ? * ) .
few objects were detected in either @xmath2 or @xmath1 , and those that were detected have low @xmath29 .
figure [ extcmd ] is a @xmath43 vs. @xmath3 color magnitude diagram for the 29 extended sources in the final catalogue .
all of the photometry has been corrected for galactic reddening . for reference ,
the point sources from figure [ cmd ] are plotted as well .
as noted previously , when viewed at higher angular resolution , many of the extended objects in the field of seyfert s sextet seen in figure [ kpnoimage ] appear to be background galaxies rather than dwarf galaxies associated with this hcg .
the photometric results for the extended sources ( figure [ extcmd ] ) lend some credence to this supposition .
the extended objects are not distributed randomly around the group , instead they appear concentrated to the northwest of the sextet . while we lack information on the southeast quadrant for comparison ( even if the pc data was analyzed , it covers significantly less area than the wf chips ) , the density of extended objects on wf3 is noticeably higher than on wf2 or wf4 .
as expected for a background cluster or group , the majority of the galaxies observed lie in a narrow region of parameter space , with @xmath71 and @xmath72 .
the six extended sources that are obvious disk galaxies are also found in this same region of the color - magnitude diagram .
in addition to the faint , red galaxies , there are three bright , red objects on wf3 ( @xmath73 and @xmath74 ) that appear to be the massive elliptical galaxy members of a poor cluster or group . based on color alone
it is difficult to say with certainty that the three bright ellipticals are giants and not dwarfs because the @xmath43 colors of dwarf galaxies and giant ellipticals can be similar . for dwarf galaxies , @xmath43 colors of @xmath75
are typical .
for example , the local group dwarf ellipticals ( des ) ngc 147 , ngc 185 , and ngc 205 have @xmath76 respectively @xcite . in a recent study of low surface brightness dwarf galaxies in the dorado group
, @xcite found 65 such objects with @xmath56 colors from @xmath77 to 2.3 .
this distribution had a strong peak at @xmath78 , again suggesting that this is a typical color for a de galaxy . for giant elliptical galaxies ( ges ) ,
the typical @xmath43 color on average may be redder by a few tenths of a magnitude than that of a typical de .
for example , @xcite observed a sample of 15 nearby ellipticals with _
hst _ and found a range of colors of @xmath79 .
although the typical colors of ges and des may only differ by a small amount , there is overlap in the overall distribution of colors between these two populations of galaxies .
thus , additional information besides the broadband colors is required to distinguish between a de and a ge .
the majority of the galaxies identified in figure [ fchart2 ] have colors redder than either the typical de or ge .
the amount of galactic reddening towards seyfert s sextet is low , and has already been removed from the colors plotted in figure [ extcmd ] .
internal , differential reddening due to dust in the sextet system is likely to affect the colors of the galaxies to some extent , but @xmath80 of @xmath81 magnitudes is required to redden dwarf galaxies with @xmath82 to @xmath83 . since the majority of the galaxies photometered are located outside of the bright isophotes of the sextet galaxies , it is unlikely that they are being reddened to such a significant extent . the assumption that the majority of the faint , red galaxies observed ( including the six background spiral galaxies , which have red @xmath43 colors ) are members of a background cluster provides a convenient explanation for the red colors of these galaxies without invoking significant internal reddening . assuming that the three bright ellipticals are giant ellipticals , then the redshift of this group is likely to be @xmath84 so that @xmath85 for these objects . at @xmath86 , the @xmath87 correction to the @xmath43 color
is @xmath90.67 magnitudes for an elliptical galaxy and @xmath90.40 magnitudes for an sa galaxy @xcite .
thus , we expect an elliptical at @xmath88 to have @xmath89 , which is similar to the colors of the ellipticals found in our sample .
perhaps the strongest evidence that these three bright , red objects are not dwarf galaxies but are instead giant galaxies comes from their surface brightness profiles . figure [ surfprof ] presents @xmath4 band surface brightness profiles for the two brightest ellipticals , which are labeled objects 3.9 and 3.15 on the finding chart ( figure [ fchart2 ] ) .
the third bright elliptical ( galaxy 3.16 ) is fainter than galaxies 3.9 and 3.15 by more than a magnitude , and its profile was not fit because the lower signal to noise surface photometry resulted in a poorly constrained fit .
the two brightest elliptical galaxies are fit well by srsic profiles of the form : @xmath90 where @xmath91 is the f814w surface brightness in mag arcsec@xmath63 and @xmath92 is fixed at @xmath93 ( i.e. , the de vaucouleurs law ) . in figure
[ surfprof ] , these @xmath94 law fits are overplotted on the surface brightness data , showing the goodness of fit . for galaxy 3.9 , the fitted central surface brightness , @xmath95 , is @xmath96 mag arcsec@xmath63 and for galaxy 3.15 , @xmath97 mag arcsec@xmath63 .
recent studies of dwarf galaxies in virgo @xcite have shown that @xmath94 law fits are inappropriate for de galaxies ; instead , the `` shape parameter '' , @xmath92 , is allowed to vary , and des show a correlation between total magnitude and @xmath92 .
the range in @xmath92 in des is found to be @xmath98 in both studies .
thus , the goodness of fit of an @xmath94 law to galaxies 3.9 and 3.15 is evidence that they are unlikely to be des .
furthermore , the central surface brightnesses of virgo des ranges from 17.8 to 24.9 in @xmath62 @xcite and from 18.1 to 25.3 in @xmath2 @xcite .
we conclude that both the shapes and central surface brightnesses of galaxies 3.9 and 3.15 are consistent with their being background ge galaxies rather than des associated with seyfert s sextet . among the sample of extended objects , there are eight galaxies with @xmath99 colors less than 1.2 , similar to the typical values for known des , and thus these objects are considered candidate dwarf galaxies .
table [ dwcand ] includes the names and photometric results for these eight objects . among these eight galaxies ,
six are simply faint ( @xmath100 ) , blue ( @xmath101 ) galaxies .
one of these six ( galaxy 2.6 ) lies near the edge of the disk of ngc6027d , while two others ( galaxies 4.1 and 4.3 ) lie near the bright tidal tail associated with ngc6027b .
galaxy 3.8 is a faint , blue galaxy that is found among the galaxies that form the suspected background cluster or group on wf3 .
however , it is found just outside of the isophotes of the tidal tail that extends from ngc 6027c .
galaxy 3.10 is also found among the suspected background cluster , but is further from the tidal feature .
galaxy 4.6 may be a background spiral , but the spiral structure is not clear enough to definitively classify this object as a disk galaxy . besides the six faint , blue dwarf galaxy candidates there are two objects with more irregular morphologies .
galaxy 4.2 is a bright , blue galaxy that shows obvious signs of star formation in the form of several compact , bright point sources superposed on a faint , -shaped continuum .
this object s morphology is reminiscent of the class of blue compact dwarfs ( see , e.g. , * ? ? ?
galaxy 3.14 is an irregularly shaped , extended object seen quite close to the disk of ngc6027c .
in addition to the color selected candidates , we include one other galaxy in table [ dwcand ] as a possible dwarf galaxy candidate .
galaxy 3.3 is one of the `` faint , red '' galaxies found on chip wf3 .
however , this galaxy is found within the boundaries of the faint tidal tail associated with ngc6027c , and it has a very peculiar morphology . this object appears to have a `` cometary '' morphology , with an elliptical galaxy at the `` head '' and a narrow `` tail '' of emission extending to the south of the head .
this galaxy appears to be a single , distorted galaxy , however , we note that a more nearby pair of galaxies exhibit a similar appearance .
the system arp 296 ( see * ? ? ? * for a @xmath2 band image ) is a face - on disk galaxy and a thin , edge - on disk galaxy that may be similar to galaxy 3.3 . if galaxy 3.3 is a system like arp 296 , then it is most likely a background object rather than a dwarf galaxy associated with the sextet .
these nine candidate dwarf galaxies stand out among the sample of extended objects because of their colors , and also , in the three specific cases described previously , because of their morphologies .
it is interesting that many of these faint galaxies lie in or near the tidal features associated with the sextet , however , this may be coincidental since the sextet galaxies fill much of the area of the three wf chips . in the end
, we refer to these galaxies as candidate dwarfs , since redshift information is necessary to discriminate between the possibilities that they are dwarf galaxies in the sextet or are background galaxies .
we hypothesized that the majority of the faint , red galaxies observed are members of a group or cluster seen in projection against seyfert s sextet .
the observations indicate that the bright , red elliptical galaxies are background giant ellipticals rather than dwarf ellipticals . however , there is little direct evidence that all of the faint , red galaxies are also background objects ; certainly the colors ( assuming a relatively significant @xmath87 correction ) and morphologies of the objects _ support _ the hypothesis that these objects are background objects , but they do not prove the hypothesis .
it is possible that some of the red galaxies may be dwarf galaxy members of seyfert s sextet , but it does seem more likely that the majority are background galaxies .
during 2002 march , spectroscopic follow - up observations of a number of galaxies identified in the _ hst _ images of seyfert s sextet were undertaken with the hobby - eberly telescope @xcite .
the observations were made with the marcario low - resolution spectrograph @xcite in multi - object mode .
three 800 second exposures were obtained and subsequently averaged together .
unfortunately , observing conditions were such that only one spectrum ( galaxy 4.2 ) was extracted with reasonable signal - to - noise .
the observations were obtained with a 600 line mm@xmath0 grism used in conjunction with @xmath102 slitlets to provide resolution of @xmath103 .
wavelength calibration for the spectrum was determined from both hgcdzn and neon comparison lamp exposures ; the resulting wavelength calibrated spectrum covers @xmath104 .
figure [ spectrum ] shows the wavelength calibrated spectrum of galaxy 4.2 over the range @xmath105 .
although a weak continuum is detected , the spectrum does contain four emission lines ; h@xmath106 and the @xmath107@xmath108 $ ] @xmath1094959 , 5007 doublet are observed at high signal - to - noise , while h@xmath110 is detected with marginal significance .
these lines and the observed wavelengths for each ( derived from gaussian fits to the line profiles ) are listed in table [ redshifts ] .
the mean redshift for the accordant galaxies in seyfert s sextet is @xmath111 km s@xmath0 @xcite .
the emission lines provide a redshift for this dwarf galaxy candidate of @xmath11220,000 km s@xmath0 , placing it in the background behind seyfert s sextet . however , it is interesting to note that this galaxy has a redshift almost identical to that of ngc6027e , the discordant spiral in the sextet , which has a redshift of 19,809 km s@xmath0 @xcite .
adopting @xmath113 for ngc 6027e and galaxy 4.2 , the distance to these two galaxies is approximately 260 @xmath114 mpc .
the absolute @xmath3 and @xmath4 magnitudes of galaxy 4.2 are approximately @xmath115 and @xmath116 respectively .
the projected separation between these two galaxies is roughly 80 , which at the adopted distance is @xmath117 kpc .
thus , we suggest that galaxy 4.2 may be a dwarf galaxy satellite of ngc6027e .
the case study of seyfert s sextet presented here is part of a continuing effort to determine whether dwarf galaxies form during tidal interactions among giant galaxies .
the sextet appears to be the most logical choice to search for tidal dwarf formation ; it is the most compact of the hickson compact groups , contains two prominent tidal tails , has a low velocity dispersion , and previous ground - based imaging revealed a number of faint , extended objects within the boundaries of the group .
however , the results of the _ hst _ imaging of seyfert s sextet show that , contrary to expectations , there is very little evidence for dwarf galaxy formation or any other strong star formation in this group .
a large number of both point sources and extended sources were catalogued and photometered from the three wide field images .
we find that very few objects are detected in either of the two blue filters , f336w and f439w , and those that are detected in the two red filters , f555w and f814w , have red colors consistent with those of old stellar populations .
the majority of the point sources detected appear to be old ( @xmath118 gyr ) and the majority of the extended sources detected appear to be background galaxies .
these photometric results contrast sharply with _
hst _ imaging studies of other hcgs , such as hcg92 ( stephan s quintet ) and hcg31 . in hcg92
, @xcite found a number of bright , blue star cluster candidates in the tidal debris regions of this group .
the images of this compact group also show bright , blue extended sources in the `` northern starburst region '' and in the tidal tails of ngc7319 and ngc7318a / b .
hcg31 contains a significant number of bright , blue point sources @xcite similar to those seen in hcg92 .
star - forming regions are also observed in hcg31 that are `` too small to be called galaxies themselves , but are not clearly associated with either galaxy ac or galaxy e '' @xcite .
thus , both hcg92 and hcg31 contain what appear to be young star clusters and tidal dwarf galaxy candidates , while seyfert s sextet does not appear to contain a significant population of either type of object .
the star cluster candidates identified in this study have photometric properties consistent with those for models of massive ( @xmath119 ) clusters with ages @xmath120 years .
the ages of these objects suggest that they are not entirely a primordial population , but may be the product of an interaction within the compact group at some time within the past few gyr .
@xcite argue that the optical tail associated with ngc 6027b and the gas that they associate with ngc 6027d may have resulted from an interaction between these two disk galaxies more than @xmath121 years ago .
the ages we derive for many of the cluster candidates are consistent with this hypothesis .
while there do not appear to be any young star clusters or tidal dwarf galaxies associated with seyfert s sextet , we did identify several candidate dwarf galaxies in the group .
this sample includes a few faint , blue extended sources , and two galaxies with peculiar morphologies : an irregularly shaped galaxy located quite near the disk of ngc6027c and an unusual , `` cometary '' galaxy located within the tidal tail associated with ngc6027c .
an additional candidate , galaxy 4.2 , has already been ruled out as a member of the sextet ; the hobby - eberly telescope spectrum of this object instead shows that it is associated with ngc6027e , the discordant redshift member of seyfert s sextet . whether or not the other candidate dwarf galaxies are associated with the sextet
, they appear morphologically very different from the clumpy , blue tidal dwarf galaxy candidates in hcg92 and hcg31 .
the data suggest that there is some fundamental , physical difference between the sextet and the two hcgs that are known to contain young star clusters and tidal dwarf galaxy candidates .
one obvious difference between these groups are the types of galaxies contained in each : seyfert s sextet is primarily made up of early - type ( s0/e ) galaxies , stephan s quintet contains spirals , and hcg31 contains mostly irregular galaxies .
based on these morphologies , one initial expectation is that the neutral gas content in seyfert s sextet is likely to be lower than that of either hcg31 or hcg92 .
radio observations show that the sextet contains only @xmath122 of neutral hydrogen @xcite , about an order of magnitude less than that of hcg31 and hcg92 @xcite .
the most recent observations of the gas content of hcg92 @xcite revise the gas mass of this group downward , however it remains at least five times larger than the gas in seyfert s sextet .
what appears to be the more significant difference among these three hcgs , however , is nt the gas mass , but the _ distribution _ of the . @xcite
present vla neutral hydrogen observations of seyfert s sextet that indicate that the majority of the mass is retained by the disk of ngc6027d , although some gas is found in a tail extending to the east of this galaxy and also in the optical tidal tail associated with ngc6027b . in hcg31 ,
the vla neutral hydrogen maps @xcite show that the gas is found both in the galaxies themselves and in a large envelope of gas that is plausibly attributed to tidal interactions between the galaxies .
the distribution of in hcg92 is found to lie entirely outside of the galaxies @xcite , however .
the is concentrated in clouds and tidal tails that are not coincident with the disks of the member galaxies .
@xcite proposed an evolutionary sequence based on their vla observations of the content of hcgs . in their model , `` phase 1 '' hcgs are those where the vast majority of the neutral gas remains bound to the member galaxies .
`` phase 2 '' hcgs are more evolved in the sense that the galaxies retain some of the gas , while approximately half of the gas mass is found in tidal features .
the final , most evolved phase is broken into two subclasses , `` phase 3a '' and `` phase 3b '' .
phase 3a groups are those where the gas is almost completely stripped from the galaxies and is found entirely within tidal features , while phase 3b groups are a few rare cases where the entire group seems to be contained in a single cloud .
hcg31 is considered a prototype phase 2 group , and hcg92 is considered an extreme example of phase 3a . both @xcite and @xcite
find seyfert s sextet to be anomalous ; its gas distribution suggests that the system has not experienced significant dynamical evolution , while optical observations suggest the opposite .
we propose one possible scenario for the history of seyfert s sextet that takes into account the following significant factors : ( 1 ) the tidal tails are evidence for interactions among the accordant redshift members some time in the past , ( 2 ) the interactions that have occurred have not triggered star and/or star cluster formation similar to that seen in other merging galaxies and compact groups , ( 3 ) the relatively small amounts of neutral gas in seyfert s sextet remains bound in the one late type galaxy and does not appear to be distributed among the group environment , and ( 4 ) the low velocity dispersion among the member galaxies and the small distances between the member galaxies suggests that future interactions among the galaxies are likely .
this accumulated evidence suggests that a number of gas - poor ( and one gas - rich ) galaxies have interacted beginning perhaps as long as 1 gyr or more in the past ( dated by the colors of the red globular cluster candidates ) .
the interactions in the group have created the optical tidal tails and perhaps created the elliptical member of the sextet , ngc6027a , as well .
the interactions in the past stripped stars from the progenitor galaxies , redistributing them within the group .
the evidence for a red , low surface brightness halo encompassing all of the member galaxies , which is seen in our images as well as deeper ground - based images , is further evidence for a redistribution of the galaxies stars within the group .
the only ongoing star formation and most of the neutral gas is found within the disk of ngc6027d , the only late type member of the group , suggesting that any interaction that involved this galaxy must have been minor , although the galaxy disk does appear somewhat irregular and perhaps warped .
we speculate that further interactions are probably inevitable , and a major interaction between ngc6027d and the other members of the group may trigger the stripping of its neutral gas and star cluster formation throughout the group in the future .
moreover , the low velocity dispersion suggests that none of the four large galaxies are likely to escape the group , and thus the group members may merge into a single galaxy , rather than remaining distinct .
thus , we believe that we are seeing seyfert s sextet at the `` beginning of the end '' ; we presume that the future interactions will be the end of this group , transforming it into a single galaxy .
we wish to acknowledge the help of m. eracleous with data reduction of our het spectra .
we also wish to thank j. hibbard for a number of useful discussions .
sz acknowledges support from the national science foundation through a research experiences for undergraduates award .
this work was supported by nasa stsci and by the nsf under grants nsf ast 00 - 71223 and stsi hst - go-08717.04-a .
lllcc alternate names : & & hcg 79 , vv115 , ngc6027 + equatorial coordinates : & & 15@xmath12359@xmath12412@xmath125 @xmath126 @xmath127 ( j2000.0 ) + mean redshift : & & 0.0145 + radial velocity dispersion : & & 138 km s@xmath0 + + galaxies : & & name & type & @xmath128 ( km s@xmath0 ) + & & ngc6027a & e0 & 4294 + & & ngc6027b & s0 & 4446 + & & ngc6027c & s0 & 4146 + & & ngc6027d & sdm & 4503 + & & ngc6027e & scd & 19809 + lrrrr 2.6 & [email protected] & [email protected] & @xmath134 & @xmath135 + 3.3 & [email protected] & [email protected] & @xmath136 & @xmath136 + 3.8 & [email protected] & [email protected] & @xmath136 & @xmath136 + 3.10 & [email protected] & [email protected] & @xmath136 & @xmath136 + 3.14 & [email protected] & [email protected] & [email protected] & [email protected] + 4.1 & [email protected] & [email protected] & @xmath137 & @xmath138 + 4.2 & [email protected] & [email protected] & [email protected] & [email protected] + 4.3 & [email protected] & [email protected] & [email protected] & [email protected] + 4.6 & [email protected] & [email protected] & @xmath136 & @xmath136 + | _ hubble space telescope _ wide field planetary camera 2 images of hickson compact group 79 , seyfert s sextet , are presented .
both point sources and extended sources detected on the three wf chips were photometered in four filters : f336w , f439w , f555w , and f814w .
unlike other hcgs that have been imaged with _ hst _ , there do not appear to be any candidate young star clusters among the detected point sources .
the majority of the point sources that may be star clusters associated with the sextet have red colors consistent with stellar populations older than 1 gyr .
a similar conclusion is drawn with regard to the extended sources .
the majority of these appear to be background galaxies , but a few candidate dwarf galaxies are identified as potentially associated with seyfert s sextet .
however , no blue , star forming objects similar to the tidal dwarf galaxy candidates identified in other hcgs are found among the extended objects identified in this study .
a redshift for one dwarf galaxy candidate was measured from a spectrum obtained with the hobby - eberly telescope , and this object was found to have a redshift similar to ngc6027e , the discordant spiral formerly identified as a member of this compact group .
the _ hst _
observations presented here and previous radio observations of the neutral gas content of this group suggest that the interactions that have taken place in the sextet only redistributed the stars from the member galaxies within the group .
we speculate that future interactions may be strong enough to strip the gas from ngc6027d and trigger star cluster formation . |
let @xmath8 be the following singularly perturbed laguerre weight @xmath9 with @xmath10 the hankel determinant is defined as @xmath11=\det(\mu_{i+j})_{i , j=0}^{k-1},\ ] ] where @xmath12 is the @xmath13-th moment of @xmath8 , namely , @xmath14 note that when @xmath15 , the integral in the above formula is convergent so that the hankel determinant @xmath16=d_k[w(x;t)]$ ] in is well - defined .
moreover , it is well - known that the hankel determinant can be expressed as @xmath17 = \prod_{j=0}^{k-1 } \gamma_{j , n}^{-2}(t);\]]see @xcite , where @xmath18 is the leading coefficient of the @xmath5-th order polynomial orthonormal with respect to the weight function in . or
, let @xmath19 be the @xmath5-th order monic orthogonal polynomial , then @xmath18 appears in the following orthogonal relation @xmath20 for fixed @xmath3 .
moreover , the monic orthogonal polynomials @xmath19 satisfy a three - term recurrence relation as follows : @xmath21with @xmath22 and @xmath23 , where the appearance of @xmath3 and @xmath4 in the coefficients indicates their dependence on @xmath3 and the parameter @xmath4 in the varying weight . in this paper , however ,
we will focus on the case when @xmath1 . since all the above integrals on @xmath24 become divergent for negative @xmath4 , we need to deform the integration path from the positive real axis to certain curves in the complex plane .
consequently , the orthogonality will be converted to the _ non - hermitian orthogonality _ in the complex plane .
more precisely , let us define the following new weight function on @xmath25 : @xmath26where @xmath27 is a complex constant , the curves @xmath28 , @xmath29 and @xmath30 ; see figure [ contour - ortho ] , @xmath31 being a positive constant .
the potential is defined in the cut plane @xmath32 $ ] as @xmath33 the orthogonality relation now takes the form @xmath34 .
, width=340 ] with the weight function @xmath35 given in , the corresponding hankel determinant @xmath36 $ ] in is well - defined .
however , since @xmath35 is not positive on @xmath37 , the orthogonal polynomials @xmath38 in may not exist for some @xmath5 , and only makes sense if all polynomials @xmath39 for @xmath40 exist .
it is worth mentioning that as part of our results , we will show that there exists a @xmath41 , such that @xmath42 exists for @xmath3 large enough and @xmath43 ; cf .
section [ sec - rhp - ops ] .
the recurrence relation still makes sense for such @xmath4 if all of @xmath44 , @xmath19 and @xmath45 exist .
note that in the literature , the polynomials with non - hermitian orthogonality have been studied in several different contexts ; see for example @xcite , where the cubic and quartic potentials are considered .
one of the main motivations of this paper comes from the wigner time - delay in the study of quantum mechanical scattering problem . to describe the electronic transport in mesoscopic ( coherent ) conductors , wigner @xcite introduced the so - called time - delay matrix @xmath46 ;
see also eisenbud @xcite and smith @xcite .
the eigenvalues @xmath47 of @xmath46 , called the proper delay times , are used to describe the time - dependence of a scattering process .
the joint distribution of the inverse proper delay time @xmath48 was found , by brouwer et al.@xcite , to be @xmath49 then the probability density function of the average of the proper time delay , namely the wigner time - delay distribution , is defined as @xmath50 the moment generating function is the laplace transformation of the wigner time - delay distribution @xmath51 which is closely related to the hankel determinant as follows : @xmath52}{d_n[w(x;0)]}.\ ] ] recently , texier and majumdar @xcite studied the wigner time - delay distribution by using a coulomb gas method .
they showed that @xmath53\}\qquad \textrm{for large } n,\ ] ] where @xmath54 is the unique minimizer for an energy problem with the external field @xmath55 in , and @xmath56 $ ] is the minimum energy .
moreover , the density @xmath57 is computed explicitly in @xcite , namely , @xmath58,~
\mbox{with}~0<a < b,~c = t/\sqrt{ab}.\ ] ] here positive @xmath59 and @xmath60 are independent of @xmath61 and implicitly determined by @xmath4 as follows : @xmath62 one may notice that @xmath54 is a probability measure on @xmath63 $ ] as long as @xmath64 is non - negative .
since @xmath64 is a continuous function of @xmath4 , we see that @xmath57 in is non - negative for @xmath65 , where @xmath66 is the critical value of @xmath4 corresponding to the case @xmath67 ; see theorem [ theorem : asymptotic of hankel ] .
it is very interesting to observe that , for this @xmath41 , we have @xmath68 and @xmath69 where a phase transition emerges at the left endpoint @xmath70 . here the critical values @xmath71 , @xmath72 and @xmath73 are explicitly given in and .
it is also interesting to look at our problem from another point of view . due to the term @xmath74 in the exponent of
, we may consider the origin as an essential singular point of the weight function . in recent years ,
orthogonal polynomials whose weights possess essential singularities have been studied extensively . for example ,
chen and its @xcite consider orthogonal polynomials associated with the weight @xmath75 they show that , for fixed degree @xmath3 , the recurrence coefficient satisfies a particular painlev iii equation with respect to the parameter @xmath4 , and the hankel determinant of fixed size @xmath76 $ ] equals to the isomonodromy @xmath6-function of the painlev iii equation with parameters depending on @xmath3 .
the matrix model and hankel determinants @xmath36 $ ] associated with the weight in were also encountered by osipov and kanzieper @xcite in bosonic replica field theories .
later , the large @xmath3 asymptotics of the hankel determinants @xmath76 $ ] associated with the weight function in is studied by the current authors in @xcite and @xcite . for @xmath77 $ ] ,
the asymptotics of the hankel determinants are derived and expressed in terms of certain painlev iii transcendents .
the asymptotics of the recurrence coefficients are also obtained therein . in the case of the gaussian weight perturbed by essential singularity @xmath78 the double scaling limit of the hankel determinants
are also characterized in terms of painlev iii transcendents by brightmore et al.in @xcite .
recently , atkin , claeys and mezzadri @xcite extend the results to the case of laguerre and gaussian weight perturbed by a pole of higher order at the origin , they obtain the double scaling asymptotics of the hankel determinants in terms of a hierarchy of higher order analogs to the painlev iii equation .
the main objective of this paper is to study the hankel determinant @xmath16 $ ] with respect to the weight in the region @xmath1 .
first , for fixed degree @xmath5 , we will show that the recurrence coefficient @xmath79 satisfies a painlev iii equation , and the hankel determinant @xmath16 $ ] equals to the isomonodromy @xmath6-function of the painlev iii equation .
then , we will derive the double scaling limit of the hankel determinant @xmath36 $ ] , the recurrence coefficients and leading coefficients of the associated orthogonal polynomials .
our results are described in terms of a certain tronque solution of the painlev i equation .
to state our results , we need certain special solutions to the painlev i equation @xmath80 the reader is referred to ( * ? ? ?
* ch.32 ) for properties of the painlev i equation , as well as the other painlev equations . in @xcite , kapaev formulates the following model riemann - hilbert ( rh , for short ) problem for @xmath81 , associated with the painlev i equation .
this model rh problem will play a crucial role later in the construction of a local parametrix in the steepest descent analysis .
* @xmath82 is analytic for @xmath83 , where @xmath84 are illustrated in figure [ contour - p1-model ] .
+ associated with the painlev i equation , width=283 ] * let @xmath85 denote the limiting values of @xmath82 as @xmath86 tends to the contour @xmath87 from the left and right sides , respectively .
then , @xmath82 satisfies the following jump conditions @xmath88 \left ( \begin{array}{cc } 1 & 0 \\ i & 1 \\ \end{array } \right ) , & z \in \gamma_k , \
k=\pm 2 ; \\[.4 cm ] \left ( \begin{array}{cc } 0 & -i\\ -i&0 \\ \end{array } \right ) , & z\in \gamma^ * , \end{array}\right .\ ] ] where @xmath89 and @xmath90 , with @xmath27 being a complex constant .
* as @xmath91 , @xmath82 satisfies the asymptotic condition @xmath92 for @xmath93 , where @xmath94 @xmath95 and @xmath96 are the pauli matrices @xmath97 it is known that , for each @xmath98 , @xmath99 is a solution of the painlev i equation . as a consequence ,
the above rh problem for @xmath82 has a solution if and only if @xmath100 is not a pole of @xmath101 . due to the meromorphic property of the painlev i transcendents ,
one also see that the solution of the above rh problem for @xmath82 is meromorphic in the parameter @xmath100 .
moreover , it is shown in kapaev @xcite that @xmath101 is the so - called _ tronque _ solution of painlev i whose asymptotic behavior is given by @xmath102\left ( 1+o(z^{-\frac 38})\right ) \ ] ] as @xmath103 and @xmath104 $ ] .
here @xmath105 is the _ tritronque _ solution satisfying @xmath106 \quad \textrm{as } z \to \infty , \ -\frac{\pi}{5}<\arg z < \frac{7\pi}{5},\ ] ] where the coefficients @xmath107 can be determined recursively ; see for example joshi and kitaev @xcite .
the solution @xmath108 will appear in our main results below .
we mention several known facts about the coefficients in in addition to .
for example , the explicit formulas of @xmath109 and @xmath110 are given in @xcite as @xmath111 where @xmath112 is the hamiltonian of painlev i. first of all , when the degree @xmath5 is fixed , we show that the recurrence coefficient @xmath113 satisfies a particular painlev iii equation with certain initial conditions .
moreover , we prove that the hankel determinant @xmath114 $ ] is related to the @xmath6-function of the painlev iii equation .
similar results for the weight in been obtained by chen and its @xcite .
[ theorem : hankel as tau function ] for fixed non - negative integer @xmath5 , let @xmath79 be the recurrence coefficient in , and @xmath115 then @xmath116 satisfies the following painlev iii equation @xmath117 with the initial conditions @xmath118 , @xmath119 . moreover , we have @xmath120=\mathrm{const } \cdot \tau(t)\ ; e^ { { n^2t}/{2}}\ ; t^ { { k(k+n)}/{2}},\ ] ] where @xmath121 is the jimbo - miwa - ueno isomonodromy @xmath6-function of the above painlev iii equation . next ,
we let @xmath7 and consider the double scaling limit when @xmath122 and @xmath123 simultaneously .
we show that the asymptotics of the hankel determinant @xmath124 $ ] associated with the weight in can be expressed in terms of the tronque solution @xmath101 of painlev i equation given in .
[ theorem : asymptotic of hankel ] let the constants @xmath71 , @xmath72 and @xmath73 be defined as @xmath125and @xmath126 for @xmath127 and @xmath128 in a way such that @xmath129 remains bounded .
suppose @xmath130 is fixed and @xmath131 is not a pole of the tronque solution @xmath101 , then an asymptotic approximation of the logarithmic derivative of the hankel determinant @xmath132 $ ] associated with the weight function is given by @xmath133 .5 cm we would also derive the double scaling limit of the recurrence coefficients and the leading coefficients of the orthonormal polynomials .
[ theorem : asymptotic of recurrence coff ] under the same conditions as in the previous theorem , the monic polynomial @xmath42 defined in exists for large enough @xmath3 and @xmath4 close to @xmath71 .
moreover , we have the asymptotics of the recurrence coefficients @xmath134 @xmath135 and @xmath136 where @xmath137 is the hamiltonian of painlev i given in .
it is well - known that the tronque solutions of painlev i are meromorphic functions and possess infinitely many poles in the complex plane . therefore , to make the results valid in the above theorems
, we require the @xmath131 in is bounded away from the poles of @xmath101 . recently , through a more delicate _
triple scaling limit _ , bertola and tovbis @xcite successfully obtain the asymptotics near the poles of @xmath101 .
similar results near the poles of @xmath101 might be derived by using their ideas in @xcite .
however , we do not pursue that part .
instead , we focus on the main task of the present paper to demonstrate that the painlev i asymptotics can also occur for the weight with negative @xmath4 . the rest of the paper is arranged as follows . in section [ finite determinants ] , we provide a rh problem for the orthogonal polynomials with respect to the weight . a transformed version of the solution is shown to fulfill a lax pair , which is closely related to the painlev iii equation .
several differential identities are stated and justified .
theorem [ theorem : hankel as tau function ] is also proved in this section
. section [ sec - e - measure ] is devoted to the determination of equilibrium measures , involving a positive measure and a signed measure . in section
[ sec - rh - analysis ] , we carry out a nonlinear steepest descent analysis of the rh problem for the orthogonal polynomials .
particular attention will be paid to the construction of the local parametrix at the critical endpoint @xmath138 , where the painlev i transcendents are involved .
then , the proofs of theorems [ theorem : asymptotic of hankel ] and [ theorem : asymptotic of recurrence coff ] are given in the last section , section [ sec - proofs ] .
in this section , we state the rh problem for the perturbed laguerre orthogonal polynomials .
then we show that after some elementary transformations , the rh problem is transformed into a rh problem for the painlev iii equation . as a consequence ,
we derive a painlev iii equation satisfied by the recurrence coefficient @xmath79 up to a translation , and establish a relation between the finite hankel determinant of the perturbed laguerre weight in with the @xmath6-function of this painlev iii equation .
several differential identities for the hankel determinants and the recurrence coefficients of the perturbed laguerre orthogonal polynomials are also derived .
the identities are important in the asymptotic analysis in later sections .
although our calculations are similar to those in chen and its @xcite , we think it is convenient for the reader to have more details .
we state the rh problem for the perturbed laguerre orthogonal polynomials as follows : @xmath139 is analytic in @xmath140 , @xmath141 ; see figure [ contour - ortho ] ; @xmath139 satisfies the jump condition @xmath142 where @xmath143 is the weight function piecewise - defined in ; the asymptotic behavior of @xmath139 at infinity is @xmath144 as @xmath145 , @xmath146 . using a by now standard argument , originally due to fokas , its , and kitaev @xcite , the solution of the above rh problem , if it exists , is uniquely given by @xmath147 -2\pi i \gamma_{k-1}^2 \;\pi_{k-1}(z ) & - \gamma_{k-1}^2\ ; \int_{\gamma } \frac { \pi_{k-1}(s ) w(s ; t ) } { s - z } ds \end{array } \right ) , \ ] ] where @xmath148 is the monic perturbed laguerre orthogonal polynomials defined in and @xmath149 is the leading coefficient for the orthonormal polynomial of degree @xmath5 . to show the existence of @xmath150 when @xmath3 is large enough
, we will apply a series of invertible transformations to transform the original rh problem @xmath151 to a new rh problem for @xmath152 , which is solvable for sufficiently large @xmath3 , @xmath4 close to @xmath71 as in , and @xmath131 is not a pole of the tronque solution @xmath101 .
tracing back the invertible transformations , we will see that the rh problem is solvable under the same conditions .
indeed , it is also possible to prove the solvability for @xmath43 .
however , since we are interested in the phase transition near @xmath71 , we do nt consider the case when @xmath153 in the subsequent analysis .
thus , the perturbed laguerre orthogonal polynomials are well - defined for @xmath7 large enough .
next , we derive some differential identities for the recurrence coefficients and the logarithmic derivative of the hankel determinant associated with the perturbed laguerre weight @xmath154 in .
the results are expressed in terms of the entries of @xmath139 .
[ lem : differential identity ] assume that @xmath155 .
let @xmath113 and @xmath156 be the recurrence coefficients in , and @xmath16 $ ] be the hankel determinant in .
define @xmath157 and @xmath158.\ ] ] then we have @xmath159 @xmath160\ ] ] and @xmath161 since @xmath155 , the orthogonal polynomials @xmath162 exist for all nonnegative @xmath5 and positive @xmath3 .
first , we consider the recurrence coefficient @xmath113 .
based on the three - term recurrence relation and the orthogonality condition , we get @xmath163 using the fact that @xmath164 and integrating by part once , the above formula gives us @xmath165 then follows from a partial fraction decomposition of @xmath166 , the orthogonality condition , and the explicit expression of @xmath139 in .
next , we consider the hankel determinant . recall that the hankel determinant can be expressed in terms of the leading coefficients as @xmath167 = \prod_{j=0}^{k-1 } \gamma_{j , n}^{-2}(t);\]]see . taking logarithmic derivative of both sides of the above equation with respect to @xmath4 and using the integral representation of the leading coefficients in , we get @xmath168 differentiating the above formula again
, we get from @xmath169 let @xmath170 be the coefficient of the @xmath171 term in @xmath38 , i.e. , @xmath172 comparing the @xmath173 powers in the recurrence relation , we obtain @xmath174 to derive @xmath175 , one can see from and that it is sufficient to obtain @xmath176 .
this can be done by taking derivative of the following orthogonal formula with respect to the parameter @xmath4 @xmath177 more precisely , taking into account the orthogonal relation and the fact that @xmath178 , we have @xmath179 then , follows from a combination of , and , as well as the definition of @xmath139 in .
finally , let us study @xmath156 . using the ideas leading to
, we have @xmath180 where @xmath170 is introduced in .
the first term on the right - hand side is @xmath181 ; cf . .
an expression for the term on the extreme right can be obtained by deriving @xmath182 from , and using and .
this completes the proof of our lemma . for later use ,
we need the differential identities of lemma [ lem : differential identity ] in the case when @xmath7 is large and @xmath183 .
they can be obtained through an analytic continuation argument .
indeed , @xmath139 determined by rh problem exists in this case , and is related to the @xmath184-function of the third painlev equation after an elementary transformation given in .
thus @xmath139 is meromorphic with respect to @xmath4 in the cut plane @xmath185 .
in particular , both @xmath151 and the hankel determinant are analytic in a domain containing the interval @xmath155 and a neighborhood of @xmath186 .
note that the identities - hold for @xmath155 , then , by analytic continuation , they also hold for @xmath4 close to @xmath71 .
we conclude that for @xmath7 large and @xmath183 , the identities - are also true .
similar argument has previously been used in bleher and deao @xcite .
introduce a purely imaginary parameter @xmath187 , and define @xmath188 where @xmath189 is the rescaled contour .
then , @xmath190 solves the following rh problem with constant jumps : * @xmath191 is analytic for @xmath192 .
as @xmath193 and @xmath37 only differ by a scale , one may refer to figure [ contour - ortho ] to see the properties of the contour @xmath193 .
* @xmath191 satisfies the jump condition @xmath194 where @xmath195 ; cf . .
* the asymptotic behavior of @xmath191 at infinity is @xmath196 where @xmath197 in the above formula , @xmath198 and @xmath199 are , respectively , the leading coefficient of the @xmath5-th orthonormal polynomial , and the coefficient of the @xmath171 term in the @xmath5-th monic orthogonal polynomial introduced in , with respect to the varying perturbed laguerre weight in and . *
the asymptotic behavior of @xmath191 at @xmath200 is @xmath201 where @xmath202 with @xmath203 now , from the above rh problem , we derive the following lax pair for the function @xmath204 , which is exactly the same as the lax pair for painlev iii ; see @xcite .
[ prop - phi - piii ] for the matrix function @xmath204 given in , we have @xmath205 where @xmath206 here , the coefficients in the above formula are given below @xmath207 @xmath208 and @xmath209 note that the jump matrices in are independent of @xmath210 and @xmath100 .
this implies that both @xmath211 are analytic functions of @xmath210 with only possible isolated singularities at the origin and at infinity . using the asymptotic expansions in - , we find that @xmath212 , \qquad a_{-2}=\frac { is}{2}\phi(0)\sigma_3\phi(0)^{-1}\ ] ] and @xmath213 , \qquad b_{-1}=-\frac { a_{-2}}{s},\]]where @xmath214=xy - yx$ ] is the commutator .
then direct computations give us the results .
it is known in several circumstances that the hankel determinants admit an interpretation as the jimbo - miwa - ueno isomonodromic @xmath6-function for the rank 2 linear system of differential equations ; see @xcite for the hankel determinants associated with the exponential weight and @xcite for more general semi - classical weights .
now we have established the relation between the perturbed laguerre orthogonal polynomials and the lax pair for the painlev iii equation .
naturally , the associated hankel determinant is also expected to relate to the @xmath6-function of the painlev iii equation .
thus we are in a position to prove our first result for fixed degree @xmath5 .
_ proof of theorem [ theorem : hankel as tau function ] .
_ according to proposition [ prop - phi - piii ] , @xmath204 satisfies the same lax pair as painlev iii .
then , applying an argument in ( * ? ? ?
* ( 5.3.4),(5.3.7 ) ) , we see that the function @xmath215 solves the painlev iii equation @xmath216 with the parameters @xmath217 and @xmath218 . by , we have @xmath219 substituting into gives us . next , we consider the hankel determinant @xmath16 $ ] .
denote by @xmath220 and @xmath221 the series in the expansions and , namely , @xmath222 with @xmath223 qiven in and @xmath224 cf .
and , where @xmath225 denotes the off - diagonal entries independent of @xmath210 . by the general theory of jimbo - miwa - ueno @xcite , the isomonodromy @xmath6-function for the lax pair in -
is defined by the formula @xmath226 where @xmath227 see ( * ? ? ?
* eq.(1.23 ) ) . substituting the definition of @xmath220 and @xmath221 into
, we obtain @xmath228 now a combination of , , and gives @xmath229see for the definition of @xmath230 and @xmath231 .
thus , we obtain from and that @xmath232 here use has been made of the relation @xmath233 . in view of the formula , and integrating both sides of , we arrive at the following relation between the hankel determinant @xmath16 $ ] and the @xmath6-function of the painlev iii equation : @xmath167=\mathrm{const } \cdot \tau(s ) e^ { { n^2t}/{2}}t^ { { k(k+n)}/{2}},\]]which is .
this completes the proof of theorem [ theorem : hankel as tau function ] .
the equilibrium measure with the external field @xmath55 in is given recently in texier and majumdar @xcite . to obtain a double scaling limit at the critical time ,
we need a modified equilibrium problem , which will involve a _ signed _ measure .
this signed measure will be used to construct the important @xmath234-function and @xmath235-function in the riemann - hilbert analysis .
the idea of considering a modified equilibrium problem has been successfully applied to study similar double scaling limits in several different problems , such as varying quartic potentials by claeys and kuijlaars @xcite and duits and kuijlaars @xcite , and a cubic potential by bleher and deao @xcite . in this section
, we will go back to the weight , consider a regular equilibrium problem first , and see how the critical time occurs . then , to facilitate our future riemann - hilbert analysis near the critical time , we will consider a modified equilibrium problem by fixing the left endpoint .
this will give us the signed measure we need .
consider the extremal problem minimizing the energy with the external field @xmath55 in : @xmath236 according to the general potential theory @xcite , there exists a unique minimizer @xmath237 of @xmath238 among all borel probability measures @xmath239 on @xmath240 , such a probability measure is called the equilibrium measure . for the potential @xmath241 in , the equilibrium measure @xmath237 can be computed explicitly .
the equilibrium measures for @xmath65 and @xmath128 have been computed explicitly in texier and majumdar @xcite . to make the present paper
self - contained , we sketch the proof below , which differs from that in @xcite . inspired by @xcite , and in view of the measure for the positive-@xmath4 case , we assume that the support of @xmath242 has only one piece . also , for fixed @xmath4 , the behavior of the density @xmath243 is expected to demonstrate a weak singularity at the endpoints since the contour is deformed to keep away from the possible singularity at the origin .
we derive the equilibrium measure by solving a scalar rh problem , based on the euler - lagrange equation .
let @xmath243 be the density function of the equilibrium measure @xmath244 supported on an interval @xmath245 , such that @xmath246 .
then for @xmath65 we have @xmath247 where @xmath248 and @xmath249 are determined by .
moreover , when @xmath250 as in , we have @xmath251 where the critical endpoints @xmath72 and @xmath73 are given in .
from , it is known that the equilibrium measure @xmath244 satisfies the euler - lagrange equation @xmath252 where @xmath253 is the lagrange multiplier . differentiating with respect to @xmath61
, we get @xmath254 where the integral is taken as the cauchy principle value .
this is an integral equation for the density function @xmath243 , which can be solved explicitly .
indeed , one can define @xmath255.\ ] ] then it follows from the plemelj formula that @xmath256 where the integral is the cauchy principle value .
it is readily verified that @xmath257 satisfies the following scalar riemann - hilbert problem : * @xmath257 is analytic for @xmath258 $ ] , having at most weak singularities at @xmath259 ; * @xmath260 for @xmath261 ; * @xmath262 as @xmath103 . solving this rh problem yields @xmath263where @xmath264 , @xmath265 and @xmath266 $ ] . here
attention should be paid to the fact that @xmath257 is analytic outside the interval @xmath267 $ ] , especially at @xmath268 .
now expanding in powers of @xmath269 , the large-@xmath2 behavior of @xmath257 ensures that @xmath270which are indeed .
furthermore , a combination of and yields . moreover , in the critical case when @xmath271 a straightforward computation gives us - .
of course , the formulas for the density function and endpoints in and hold when @xmath272 . for any @xmath15 , the density function is supported on @xmath63 $ ] with @xmath273 and vanishes like square roots at both endpoints . as a consequence ,
one will obtain usual airy - type and sine - type asymptotic expansions for the orthogonal polynomials near the endpoints and inside the support , respectively .
the critical case when @xmath186 is termed a _ freezing transition _ ; see @xcite .
one can see that , when @xmath186 , the density function @xmath243 in vanishes like a @xmath274 root at @xmath72 .
this suggests that the local behavior of the orthogonal polynomials near @xmath72 is described in terms of the painlev i transcendents ; see @xcite . to precisely construct a local parametrix near the endpoint @xmath72 by using the painlev i transcendents , a delicate study near @xmath72
is needed in our subsequent nonlinear steepest descent analysis for the rh problem .
therefore , technically it is more convenient to have a measure whose left endpoint of the support is exactly located at @xmath72 .
note that in the case when only positive measures are involved as in section [ sec - e - measure - regular ] , both endpoints @xmath59 and @xmath60 in vary when the value of parameter @xmath4 changes .
so we need to minimize the same energy functional as there is no symmetry as in @xcite , the right endpoint @xmath60 may depend on @xmath4 .
a similar treatment is also employed in @xcite . by a similar argument performed in section [ sec - e - measure - regular ] , we find the new minimizer explicitly .
let @xmath275 be the signed density function of the minimizer of the minimizer of the energy functional in .
then we have @xmath276,\ ] ] where @xmath277 and @xmath60 is determined by the equation @xmath278 .5 cm it is worth noting that for @xmath186 , the density of the modified equilibrium measure @xmath275 is reduced to @xmath279 in .
moreover , near the critical time @xmath186 , we have @xmath280 & d_0=a_{cr}^2-\sqrt{\frac{a_{cr}}{b_{cr } } } \frac{2b_{cr } - a_{cr}}{2(b_{cr } - a_{cr})}(t - t_{cr})+o\left ( ( t - t_{cr})^2\right ) , \\[.2 cm ] & d_1=-2a_{cr}+\frac 12\sqrt{\frac{b_{cr}}{a_{cr}}}(b_{cr}-a_{cr})^{-1}(t - t_{cr})+o\left ( ( t - t_{cr})^2\right ) .
\end{array}\ ] ] based on the signed measure obtained above , we define several auxiliary functions which will be used in our further analysis .
the @xmath234-function is defined as @xmath281,\ ] ] where @xmath282 , and the equilibrium density function @xmath275 is given in .
we also define the following @xmath235-functions @xmath283 where the branches are chosen such that @xmath284 , @xmath285 and @xmath286 . from the above definitions
, it is immediately seen that @xmath287 satisfies the euler - lagrange equation @xmath288 and the variational inequality @xmath289where @xmath253 is the lagrange multiplier introduced in .
moreover , the @xmath234-function and the @xmath235-function are related by @xmath290 note that @xmath291 and @xmath292 are close to each other when @xmath4 approaches @xmath71 .
if we rewrite @xmath291 as @xmath293 then , in view of - , we have @xmath294 where @xmath295 is analytic in a neighborhood of @xmath138 and @xmath296 .
we also need some local information of the functions @xmath292 and @xmath291 at critical points @xmath138 and @xmath268 . from their definitions in and
, we have @xmath297 where @xmath298 , and @xmath299 where @xmath300 . from the above formula
, one can see that @xmath301 if @xmath1 and @xmath2 approaches the origin such that @xmath302 ; see figures [ contour - phi - two ] and [ contour - phi - cr ] .
in particular , we see that @xmath303 is exponentially small as @xmath145 , @xmath304 or @xmath305 ; cf . figure [ contour - ortho ] for the contours .
in view of , one can see that the same holds for @xmath306 .
it is worth mentioning that on @xmath307 and @xmath308 with @xmath309 , we have @xmath310 . .
the left and right pictures correspond to cases when @xmath1 and @xmath155 , respectively.,width=585 ] .
note that this figure is not the exact one for @xmath311 defined in .
here we have rescaled the figure , especially near @xmath72 , for better illustration : because the exact value of @xmath72 is too small as compared with @xmath73 ; see .
, width=415 ]
in this section , we apply the nonlinear steepest descent method developed by deift and zhou et al.@xcite to the rh problem for @xmath151 . the idea is to obtain , via a series of invertible transformations @xmath312 , the rh problem for @xmath152 whose jump matrices are close to the identity ones .
we make use of the @xmath234-function defined in to normalize the rh problem for @xmath151 in section [ sec - rhp - ops ] when @xmath7 . as @xmath314 for large @xmath315
, we introduce the first transformation @xmath316 as follows : @xmath317 where @xmath253 is the lagrange multiplier in .
then , @xmath318 solves the following rh problem .
( t1 ) @xmath319 is analytic in @xmath320 ; see figure [ contour - ortho ] for @xmath25 ; ( t2 ) the jump condition is @xmath321 for @xmath322 , @xmath141 , where @xmath323 is defined in , and @xmath324 , @xmath325 , @xmath326 ; ( t3 ) the asymptotic behavior of @xmath319 at infinity is @xmath327 appealing to the properties of @xmath287 and @xmath328 in and , the jump matrices in can be expressed in terms of the function @xmath328 as follows : @xmath329 \left ( \begin{array}{cc } e^{2n\left ( \phi_t\right ) _ + ( z ) } & 1 \\ 0 & e^{2n\left ( \phi_t\right ) _
-(z ) } \\ \end{array }
\right ) , & z\in(a_{cr},b ) .
\end{array } \right .\ ] ] since @xmath331 are purely imaginary on @xmath332 , the jump matrix for @xmath319 on @xmath333 possesses highly oscillatory diagonal entries . to remove the oscillation , we open the lens near @xmath332 and introduce the second transformation : @xmath334 t(z ) \left ( \begin{array}{cc } 1 & 0 \\ -e^{2n\phi_t(z ) } & 1 \\ \end{array } \right ) , & \mbox{for $ z$ in the upper lens region;}\\[.4 cm ] t(z ) \left ( \begin{array}{cc } 1 & 0 \\ e^{2n\phi_t(z ) } & 1 \\ \end{array } \right ) , & \mbox{for $ z$ in the lower lens region . }
\end{array}\right .\ ] ] .
the shaded region is the region where @xmath335.,width=453 ] then @xmath336 solves the rh problem ( s1 ) @xmath336 is analytic in @xmath337 ; see figure [ contour - s ] for the contours ; ( s2 ) the jump conditions are @xmath338 \left ( \begin{array}{cc } 0 & 1 \\
-1 & 0 \\ \end{array } \right ) , & z\in(a_{cr},b ) , \\ [ .4 cm ] \left ( \begin{array}{cc } 1 & \alpha e^{-2n \phi_t(z ) } \\ 0 & 1 \\ \end{array } \right ) , & z\in \gamma_2,\\[.4 cm ] \left ( \begin{array}{cc } 1 & ( 1-\alpha)e^{-2n \phi_t(z ) } \\ 0 & 1 \\ \end{array } \right ) , & z\in \gamma_3,\\[.4 cm ] \left ( \begin{array}{cc } 1 & e^{-2n \phi_t(z ) } \\ 0 & 1 \\ \end{array}\right ) , & z\in ( b,+\infty ) .
\end{array}\right.\ ] ] ( s3 ) the asymptotic behavior at infinity is @xmath339 to study the asymptotic behavior of @xmath336 for large @xmath3 , we may exam the signs of @xmath340 , to see if the jumps are of the form @xmath341 plus exponentially small terms .
special attention should be paid in the present case since we are dealing with the signed measure .
fortunately , when @xmath3 is large and @xmath342 is small enough , we can still determine the signs of @xmath340 near the endpoint @xmath72 .
similar discussions can be found in @xcite where modified equilibrium problems are also addressed .
[ prop - phi - sign ] let @xmath343 be a neighbourhood of @xmath72 .
then , for any @xmath344 , there exists a @xmath345 such that for all @xmath346 with @xmath347 , we have @xmath348 on the upper and lower lips of the lens on the outside of @xmath343 , namely , @xmath349 .
moreover , there exists a positive @xmath350 , such that @xmath351 on @xmath352 and on @xmath353 . in view of
, we see that the factor @xmath354 in possesses a pair of zeros @xmath355 one can choose @xmath356 small enough , so that @xmath357 .
similar to those conducted in ( * ? ?
* prop.4.2 ) and @xcite , we can prove that the jumps on the portions of @xmath358 and @xmath359 , outside of @xmath343 and keeping a distance from the soft edge @xmath60 , are of the form @xmath341 plus an exponentially small term .
next , we estimate @xmath340 on @xmath360 in a straightforward manner , using the explicit representations of the @xmath235-functions , and . in view of
, we need only check the critical case for @xmath311 , and it is readily seen from that @xmath361 for @xmath2 on the semi - circle @xmath307 ; cf . figure [ contour - ortho ] , we take the integration path to be the arc from @xmath72 to @xmath2 , and adapt the parametrization @xmath362 , where @xmath363 and @xmath364 . as a result , we have @xmath365 where @xmath366 such that @xmath367 .
what is more , we have @xmath368\subset [ 3\pi/4 , \pi]$ ] on @xmath307 , where @xmath369 so that @xmath370 .
hence the argument of the integrand lies in the interval @xmath371 , so long as @xmath372 .
therefore , from we see that @xmath373 , and @xmath374 is strictly monotonically increasing as @xmath304 goes away from @xmath72 .
the same result holds for @xmath305 .
it is worth mentioning that @xmath375 as @xmath145 , @xmath376 ; see the formula and the discussion that follows .
we note that the estimates on the lens boundaries @xmath358 and @xmath359 can also be obtained from the representations , and .
having had proposition [ prop - phi - sign ] , we see from that the jump matrix for @xmath377 is the identity matrix , plus an exponentially small term , for fixed @xmath2 bounded away from the interval @xmath332 . neglecting the exponentially small terms , we arrive at an approximating rh problem for @xmath378 as follows : ( n1 ) @xmath378 is analytic in @xmath379 $ ] ; ( n2 ) @xmath380 ( n3 ) @xmath381 the solution to the above rh problem is constructed explicitly as @xmath382 where @xmath383 and @xmath384 with @xmath385 and @xmath386 .
the jump matrices of @xmath387 are not uniformly close to the identical matrix @xmath341 near the endpoints @xmath72 and @xmath60 , thus local parametrices have to be constructed in neighborhoods of these endpoints . the local parametrix at the right endpoint @xmath388 is the same as that of the laguerre polynomials at the soft edge .
more precisely , the parametrix is to be constructed in @xmath389 , @xmath350 being a fixed positive number , such that ( a ) @xmath390 is analytic in @xmath391 ; see figure [ contour - s ] for the contour @xmath392 ; ( b ) in @xmath393 , @xmath390 satisfies the same jump conditions as @xmath336 does ; cf . ; ( c ) @xmath390 fulfils the following matching condition on @xmath394 : @xmath395 the parametrix can be constructed , out of the airy function and its derivative , as in ( * ? ? ?
* ( 3.74 ) ) ; see also @xcite .
now we focus on the construction of the parametrix at the endpoint @xmath72 .
we seek a parametrix in @xmath396 , @xmath350 being a fixed positive number , such that the following rh problem is satisfied : ( a ) @xmath397 is analytic in @xmath398 ; cf .
figure [ contour - s ] ; ( b ) in @xmath399 , @xmath397 satisfies the same jump conditions as @xmath336 does ; cf .
; ( c ) @xmath397 fulfils the following matching condition on @xmath400 : @xmath401 first we make a transformation to convert all the jumps of the rh problem for @xmath397 to constant jumps .
let us define @xmath402 then it is readily verified that @xmath403 satisfies a rh problem as follows : ( a ) @xmath404 is analytic in @xmath398 ; ( b ) in @xmath405 , @xmath404 satisfies the jump conditions @xmath406 \left ( \begin{array}{cc } 1 & 1- \alpha \\ 0 & 1 \\ \end{array } \right ) , & z \in \gamma_3 , \\[.4 cm ] \left ( \begin{array}{cc } 0 & 1\\ -1&0 \\
\end{array } \right ) , & z\in ( a_{cr},a_{cr}+r ) , \\[.4 cm ] \left ( \begin{array}{cc } 1 & 0 \\ 1 & 1 \\ \end{array } \right ) , & z \in \gamma_4 \cup \gamma_5 .
\end{array}\right .\ ] ] we are now in a position to construct a solution for the above rh problem by using the @xmath184-function associated with the painlev i equation , as introduced in section [ sec - model riemann - hilbert problem ] .
we note that @xmath407 shares the same jumps with the @xmath184-function .
then , we establish the following conformal mapping : @xmath408and @xmath350 being sufficiently small ; cf . . indeed , in view of
, one can improve to obtain @xmath409 as @xmath410 .
moreover , we have @xmath411 where the fractional power takes the principle branch .
we also define @xmath412 where the square root takes the principal branch again .
it follows from the definitions of @xmath413 and @xmath414 in and that @xmath415 is analytic in a neighborhood of @xmath138 and @xmath416 moreover , we have @xmath417 where @xmath418 is the function defined in . with all these preparations ,
the parametrix near the endpoint @xmath72 can be constructed explicitly as @xmath419 where @xmath420 is defined as @xmath421 it is ready to see that @xmath420 is analytic in @xmath405 and the function @xmath397 in indeed satisfies the matching condition .
[ rmk - matching ] as we have discussed in section [ sec - model riemann - hilbert problem ] , the solution @xmath82 exists if and only if @xmath100 is not a pole of @xmath101 .
then , to make @xmath397 in well - defined , we choose @xmath422 as @xmath122 , and require @xmath423 is not a pole of @xmath101 . moreover , from the large-@xmath86 behavior of @xmath82 in , we can verify the desired matching condition on @xmath400 in .
with all the parametrices constructed , let us introduce the final transformation @xmath425 s(z ) ( p^{(a)})^{-1}(z ) , & z\in u(a , r)\backslash \sigma_{s } ; \\[.1 cm ] s(z ) ( p^{(b)})^{-1}(z ) , & z\in u(b , r)\backslash \sigma_{s } .
\end{array}\right .\ ] ] then @xmath426 satisfies a rh problem as follows : ( r1 ) @xmath426 is analytic in @xmath427 ; see figure [ contour - r ] ; ( r2 ) @xmath426 satisfies the jump conditions @xmath428 where @xmath429 p^{(b)}(z)n^{-1}(z ) , & z\in\partial u(b , r),\\[.1 cm ] n(z)j_s(z)n^{-1}(z ) , ~ & \sigma_r\setminus \partial ( u(a_{cr},r)\cup u(b , r ) ) ; \end{array}\right .\ ] ] ( r3 ) @xmath426 satisfies the following behavior at infinity : @xmath430 , width=377 ] based on the matching conditions , and the properties of the function @xmath291 stated in proposition [ prop - phi - sign ] , we have the following estimates : @xmath431 i+o\left ( \frac 1n\right ) , & z\in\partial u(b , r),\\[.1 cm ] i+o(e^{-cn } ) , ~ & z\in \sigma_r\setminus ( \partial u(a , r)\cup \partial u(b , r ) ) , \end{array}\right .\ ] ] where @xmath432 is a positive constant , and the error term is uniform for @xmath2 on the corresponding contours .
here we also require that @xmath433 ; see remark [ rmk - matching ] . then , applying the standard argument of integral operator and using the technique of deformation of contours , see for example @xcite , we can see that @xmath426 exits when @xmath3 is large enough and @xmath4 is close to @xmath71 .
moreover , we have @xmath434 uniformly for @xmath2 in the whole complex plane .
this completes the nonlinear steepest descent analysis .
[ sec - proofs ] to prove our main results , theorems [ theorem : asymptotic of hankel ] and [ theorem : asymptotic of recurrence coff ] , we need more refined asymptotic approximations for @xmath426 than the one we get in . for this purpose , we derive an asymptotic approximation for @xmath426 . to simplify the statement of our result
, we denote @xmath435 let @xmath436 and @xmath437 ; see . for @xmath127 , @xmath438 as in , and @xmath131
is not a pole of the tronque solution @xmath101 , we have @xmath439 where @xmath440 and @xmath441 the constants @xmath442 and @xmath443 in the above formulas are given as @xmath444and @xmath445 , cf . . from and
, we derive an asymptotic expansion for the jump @xmath446 @xmath447 where @xmath448 for @xmath449 , @xmath450 is expanded in terms of @xmath451 for non - negative integers @xmath5 .
thus , we have for large @xmath3 .
to derive the explicit formulas of @xmath452 and @xmath453 , we combine , with . then one can see that @xmath452 and @xmath453 satisfy rh problems as follows : * @xmath452 and @xmath453 are analytic for @xmath454 ; * @xmath452 and @xmath453 satisfy the jump conditions @xmath455 and @xmath456 * as @xmath457 , both @xmath452 and @xmath453 are of order @xmath458 .
it follows from the plemelj formula that @xmath452 and @xmath453 can be represented as the cauchy - type integrals on the circular contour @xmath400 of the right - hand terms in and , respectively .
the above rh problems can then be solved by conducting residue calculations of the cauchy - type integrals . as a direct consequence
we obtain and .
now we are ready to derive the large-@xmath3 asymptotics for the recurrence coefficients and the hankel determinant , as stated in our main theorems , theorems [ theorem : asymptotic of hankel ] and [ theorem : asymptotic of recurrence coff ] .
_ proof of theorems [ theorem : asymptotic of hankel ] and [ theorem : asymptotic of recurrence coff ] .
_ tracing back the transformations @xmath459 in , and , we have @xmath460 for @xmath2 close to the origin .
to apply the differential identities in lemma [ lem : differential identity ] , we need to extract the asymptotics of @xmath461 when @xmath7 , @xmath127 , @xmath438 as in , and @xmath131 is not a pole of the tronque solution @xmath101 . from the explicit formula of @xmath378 in and the approximation of @xmath426 in
, we have @xmath462 where @xmath463 and the constants @xmath464 are defined in . now in lemma [ lem : differential identity ] implies that @xmath465 then the asymptotic formula for the hankel determinant follows immediately from the identities and .
this completes the proof of theorem [ theorem : asymptotic of hankel ] . to derive the asymptotics of the recurrence coefficient @xmath466 and the leading coefficient @xmath467
, we use the relations @xmath468 where @xmath469 is the coefficient of the @xmath470 term in the asymptotic expansion of @xmath471 , that is , @xmath472 therefore , we also need to expand @xmath378 and @xmath426 as @xmath473 again , using the explicit formula of @xmath378 in and the asymptotic approximation of @xmath426 in , we have @xmath474 and @xmath475 where @xmath476 recalling the values of @xmath442 and @xmath443 in , and
the identities @xmath477 and @xmath478 , we have from and that @xmath479this is . similarly , the asymptotics for the leading coefficient can be obtained .
indeed , we have @xmath480 which gives us . finally , we derive the asymptotics for @xmath481 .
note that , from , we have @xmath482 to find the asymptotic behavior of @xmath483 , we have from and that @xmath484 then , the asymptotic approximation for @xmath481 in follows from the above formula and .
this completes the proof of theorem [ theorem : asymptotic of recurrence coff ] .
the authors are grateful to the referee for the valuable comments and suggestions to improve the rigorousness of the paper .
the work of shuai - xia xu was supported in part by the national natural science foundation of china under grant numbers 11201493 and 11571376 , guangdong natural science foundation under grant numbers s2012040007824 and 2014a030313176 , and the fundamental research funds for the central universities under grant number 13lgpy41 .
dan dai was partially supported by grants from the research grants council of the hong kong special administrative region , china ( project no .
cityu 11300115 , 11300814 ) .
yu - qiu zhao was supported in part by the national natural science foundation of china under grant numbers 10871212 and 11571375 .
m. bertola and a. tovbis , asymptotics of orthogonal polynomials with complex varying quartic weight : global structure , critical point behaviour and the first painlev equation , _ constr .
_ , * 41*(2015 ) , 529 - 587 .
p. deift , t. kriecherbauer , k.t .- r .
mclaughlin , s. venakides and x. zhou , uniform asymptotics for polynomials orthogonal with respect to varying exponential weights and applications to universality questions in random matrix theory , _ comm .
pure appl .
_ , * 52*(1999 ) , 1335 - 1425 .
p. deift , t. kriecherbauer , k.t .- r .
mclaughlin , s. venakides and x. zhou , strong asymptotics of orthogonal polynomials with respect to exponential weights , _ comm .
pure appl .
_ , * 52*(1999 ) , 1491 - 1552 .
fokas , a.r .
its , a.a .
kapaev and v.yu .
novokshenov , _ painlev transcendents : the riemann - hilbert approach _ , ams mathematical surveys and monographs , vol .
128 , amer .
soc . , providence r.i . , 2006 .
m. jimbo , t. miwa and k. ueno , monodromy preserving deformation of linear ordinary differential equations with rational coefficients .
i. general theory and @xmath6-function , _ phys .
d _ , * 2*(1981 ) , 306 - 352 . | in this paper , we consider polynomials orthogonal with respect to a varying perturbed laguerre weight @xmath0 for @xmath1 and @xmath2 on certain contours in the complex plane . when the parameters @xmath3 , @xmath4 and the degree @xmath5 are fixed , the hankel determinant for the singular complex weight is shown to be the isomonodromy @xmath6-function of the painlev iii equation .
when the degree @xmath7 , @xmath3 is large and @xmath4 is close to a critical value , inspired by the study of the wigner time delay in quantum transport , we show that the double scaling asymptotic behaviors of the recurrence coefficients and the hankel determinant are described in terms of a boutroux tronque solution to the painlev i equation .
our approach is based on the deift - zhou nonlinear steepest descent method for riemann - hilbert problems .
2010 _ mathematics subject classification _ : primary 33e17 , 34m55 , 41a60 . _ keywords and phrases _ : asymptotics ; hankel determinants ; painlev i equation ; painlev iii equation ; riemann - hilbert approach . |
one of the strongest motivations to investigate magnetic semiconductors and half - metallic ferromagnets ( hmf ) is the possibility to design and produce novel stable structures on semiconducting substrates with new interesting properties . adopting this point of view first - principle studies
are an excellent starting point to predict new systems having the desired properties .
recently akinaga _ et al .
_ @xcite found the possibility to fabricate zinc - blende ( zb ) type cras half - metallic ferromagnetic material .
experimental data confirmed that this material is ferromagnetic with the magnetic moment of @xmath0 , in agreement with theoretical predictions @xcite .
according to this calculation this half - metallic material has a gap of about @xmath1 in the minority spin channel which has attracted much attention to this potential candidate for spintronic applications , keeping in mind also its high curie temperature @xmath2 around @xmath3 .
note that recent experiments on cras epilayers grown on gaas@xmath4 evidenced an orthorhombic structure , different from the zb one , so the structure is rather sensitive to the preparation process @xcite .
however , it is highly desirable to explore the possibility of existence of half - metallic ferromagnetism in materials which are compatible with important iii - v and ii - iv semiconductors . for this purpose efforts
has been made on the metastable zb structures , such as cras @xcite . beyond the importance of the preparation techniques
, attention should be given to the understanding of the finite temperature properties in these hmf materials .
therefore , it is interesting to explore theoretically the mechanism behind half - metallic ferromagnetism at finite temperature from a realistic electronic structure point of view .
theoretical studies @xcite of the @xmath5 transition metal monoarsenides have shown that the ferromagnetic phase of zb structure cras compound should be more stable than the antiferromagnetic one .
the calculated equilibrium lattice constant is larger than that of gaas ( @xmath6 ) having a value of @xmath7 @xcite . following this work similar electronic structure calculations concerning the stability of the half - metallic ferromagnetic state in the zb structure
have been carried out @xcite .
however standard local density approximation lda ( or gga , lda+u , etc . )
calculations are in general insufficient to describe some important many - particle features of the half - metallic ferromagnets .
one of these many - body features , the non - quasiparticle ( nqp ) states @xcite contribute essentially to the tunneling transport in heterostructures containing hmf @xcite .
the origin of these states is connected with the `` spin - polaron '' processes : the spin - down low - energy electron excitations , which are forbidden for hmf in the one - particle picture , turn out to be possible as superpositions of spin - up electron excitations and virtual magnons @xcite .
recently we have applied the lda+dmft ( dynamical mean field theory ) approach @xcite to describe from first principles the non - quasiparticle states in a prototype half - metallic ferromagnet nimnsb @xcite . in this paper
, we describe the correlation effects in cras hmf material in the framework of the lda+dmft approach .
we will show that these many - body spin - polaron processes are very sensitive to structural properties of the artificially produced cras compound . depending on the substrate characteristics , such as a large lattice constant , as in the case of inas
( @xmath8 ) , or a smaller one as in the case of gaas ( @xmath6 ) the spectral weight of the non - quasiparticle states can be tuned from a large value in the former case ( inas ) to an almost negligible contribution in the later case ( gaas ) .
therefore , the correlation effects in conjunction with structural properties determine the behavior of electronic states in cras near the fermi level which has a substantial impact on the tunneling transport in the corresponding heterostructures .
first , let us explain in a simple way why the non - quasiparticle states are important for the tunneling transport ( for formal derivations , see refs . ) . to this aim
we consider the case of a narrow - band saturated hubbard ferromagnet where the current carriers are the holes in the lowest hubbard band and the non - quasiparticle states provide _ all _ spectral weight for the minority spin projection @xcite . a schematic density of states is shown in fig .
[ model]a .
suppose we have a tunnel junction with two pieces of this ferromagnet with either parallel ( fig .
[ model]b ) or antiparallel ( fig .
[ model]c ) magnetization directions . from the one - particle point of view
, the spin - conserving tunneling is forbidden in the latter case . however , in the framework of many - particle picture the charge current is a transfer process between an empty site and a single - occupied site rather than the motion of the electron irrespective to the site like in the band theory and therefore the distinction between these two cases ( see fig . [ model ] ) , is due only to the difference in the densities of states .
it means that the estimations of the tunneling magnetoresistance based on simple one - electron picture is too optimistic ; even for antiparallel spin orientation of two pieces of the half - metallic ferromagnets in the junction for zero temperature the current is not zero , due to the non - quasiparticle states .
more exactly , it vanishes for zero bias since the density of nqp states at the fermi energy equals to zero .
however , it grows with the bias sharply , having the scale of order of typical _ magnon _ energies , that is , millivolts . the latter statement
is confirmed by a formal consideration of the antiparallel spin orientation case , based on the standard tunneling hamiltonian @xcite : @xmath9 where @xmath10 are the hamiltonians of the left ( right ) half - spaces , @xmath11 and @xmath12 are corresponding quasimomenta and spin projections are defined with respect to the magnetization direction of a given half - space with opposite magnetization directions ( spin is supposed to be conserving in the `` global '' coordinate system ) . for the tunneling current @xmath13 in the second order in @xmath14 one has @xmath15 \nonumber \\ & \times & \left [ f(t_k)-f(t_k+ev)\right ] \delta(ev+t_k - t_{p - q}+\omega_q)\end{aligned}\ ] ] where @xmath16 is the electron charge and @xmath17 is the bias @xcite .
the differential conductance of the junction with antiparallel magnetization directions is just proportional ( for zero temperature ) to the density of the non - quasiparticle states , @xmath18 note that the value @xmath19 vanishes for @xmath20 , where @xmath21 is an anisotropy gap in the magnon spectrum @xcite , which is small , but could be changed by suitable substitution @xcite .
it is worthwhile to note also that the nonquasiparticle states play an important role ( together with the classical effect of the smearing of the gap by spin - disorder scattering ) in the depolarization of the states near the fermi energy at finite temperatures ; again , estimations of the tunneling magnetoresistance based on a simple stoner - like picture of the energy spectrum appear to be completely unrealistic @xcite .
to perform the calculations we used the newly developed lda+dmft scheme @xcite .
this method is based on the so - called exact muffin - tin orbitals ( emto ) scheme @xcite within a screened kkr @xcite approach , frozen core together with the local spin density approximation ( lda ) .
the correlation effects are treated in the framework of dynamical mean field theory ( dmft ) @xcite , with a spin - polarized t - matrix fluctuation exchange ( sptf ) type of dmft solver @xcite .
the sptf approximation is a multiband spin - polarized generalization of the fluctuation exchange approximation ( flex ) @xcite , but with a different treatment of particle - hole ( ph ) and particle - particle ( pp ) channels .
the particle - particle ( pp ) channel is described by a @xmath22-matrix approach @xcite giving a renormalization of the effective interaction .
this effective interaction is used explicitly in the particle - hole channel .
justifications , further developments and details of this scheme can be found in ref . .
the spin - polaron process is described by the fluctuation potential matrix @xmath23 with @xmath24 , defined in a similar way as in the spin - polarized flex approximation @xcite : @xmath25 the essential feature here is that the potential ( [ w ] ) is a complex energy dependent matrix in spin pace with _ off - diagonal _ elements : @xmath26 where @xmath27 represents the bare vertex matrix corresponding to the transverse magnetic channel , @xmath28 is an effective transverse susceptibility matrix and @xmath29 is the bare transverse susceptibility @xcite .
@xmath30 are fermionic matsubara frequencies and @xmath31 corresponds to the magnetic interaction channel @xcite . the local green functions as well as
the electronic self - energies are spin diagonal for collinear magnetic configurations . in this approximation
the electronic self - energy is calculated in terms of the effective interactions in various channels .
the particle - particle contribution to the self - energy was combined with the hartree - fock and the second order contributions @xcite . to ensure a physical transparent description in the current implementation
the combined particle - particle selfenergy is presented by its a hartree @xmath32 and fock @xmath33 types of contributions : @xmath34 where the particle - hole contribution @xmath35 reads : @xmath36 since some part of the correlation effects are included already in the local spin - density approximation ( lsda ) `` double counted '' terms should be taken into account .
to this aim , we start with the lsda electronic structure and replace @xmath37 by @xmath38 in all equations of the lda+dmft method @xcite , the energy @xmath39 being relative to the fermi energy and @xmath40 .
it means that we only add _ dynamical _ correlation effects to the lsda method . in our calculations we considered the standard representation of the zinc - blende structure with an fcc unit cell containing four atoms : cr@xmath41 ,
as@xmath42 and two vacant sites e@xmath43 and e1@xmath44 .
we used the experimental lattice constant of gaas(@xmath45 ) and inas(@xmath46 ) for all the calculations , and the equilibrium value for the bulk cras ( a@xmath47 ) , obtained by flapw @xcite . to calculate the charge density we integrate along a contour on the complex energy plane which extends from the bottom of the band up to the fermi level @xcite , using 30 energy points . for the brillouin zone integration
we sum up a k - space grid of 89 points in the irreducible part of the brillouin zone .
a cutoff of @xmath48 for the multi - pole expansion of the charge density and a cutoff of @xmath49 for the wave functions was used .
the perdew - wang @xcite parameterization of the lda to the exchange correlation potential was used .
the lda+dmft calculations were carried out for three lattice constants : the gaas , inas and the `` equilibrium '' one .
the corresponding lda computational results agree with previous ones @xcite . in order to evaluate the average coulomb interaction on the cr atoms and the corresponding exchange interactions we start with the constrained lda method @xcite . in our case
the constrained lda calculation indicates that the average coulomb interaction between the cr @xmath5 electrons is about @xmath50 ev with an exchange interaction energy about @xmath51 ev .
it is important to note that the values of the average coulomb interaction parameter slightly decrease going from the gaas ( @xmath52 ev ) to inas ( @xmath53 ev ) lattice constants , see table tab .
this is in agreement with a naive picture of a less effective screening due to increasing of the distances between the atoms .
the typical insulating screening used in the constraint calculation @xcite should be replaced by a metallic kind of screening .
the metallic screening will lead to a smaller value of @xmath54 .
unfortunately , there are no reliable schemes to calculate @xmath54 in metals , therefore we choose some intermediate value of @xmath55 ev and @xmath51 ev .
it is important to realize that there are no significant changes in the values of the average coulomb interaction for the studied lattice structures , and that the exchange interaction is practically constant .
note that physical results are not very sensitive to the value of @xmath54 , as it was demonstrated by us for nimnsb @xcite .
the lsda and lsda+dmft calculation for the density of states , dos , is presented in fig .
[ 200_cras_dos ] . depending on the lattice constant the cr and as atoms loose electrons and this charge
is gained by the vacant sites . as a result
the fermi level is moving from the right edge of the gap as for the case of gaas substrate , towards the middle of the gap for a inas substrate .
the cr moments are well localized due to a mechanism being similar to that of localization of the magnetic moment on the mn atom in the heusler nimnsb @xcite .
note that the local cr spin moment is more than @xmath56 , the as atom possesses a small induced magnetic moment ( of order of @xmath57 ) coupled antiparallel to the cr one .
the results are presented in table [ tab2 ] .
calculated , dmft , cr magnetic moments increased in comparison with the lda results , due to the localization tendency of the cr@xmath58 states as a consequence of correlation effects @xcite .
according to our calculations , the system remains half - metallic with a pretty large band gap ( about @xmath59 ev ) for all the lattice constants compared with the band gap of the nimnsb which is only @xmath60 ev @xcite . in fig .
[ 200_cras_dos ] the non - quasiparticle states are visible for lattice parameter higher than the equilibrium one , with a considerable spectral weight in the case of the inas substrate .
inas is widely used as substrate material for the growth of different compounds . according to our calculations
this situation is very favorable for the experimental investigation of the formation of non - quasiparticle states as well as the investigation of their nature . comparing with the case of the gaas lattice parameter , for the inas substrate
, the most significant change in the electronic structure is suffered by the as-@xmath12 states . having a larger lattice constant the cr atom acquires a slightly larger magnetic moment . in the lda calculations , nevertheless , the magnetic moment per unit cell is integer @xmath56 . expanding the lattice constant from gaas to inas lattice
the cr states become more atomic like and therefore the spin magnetic moment increases .
this is reflected equally in the charge transfer which is smaller for the inas lattice parameters .
a larger cr moment induces a large spin polarization of the as-@xmath12 states , compensating the smaller @xmath61 hybridization , the total moment remaining at its integer value of @xmath56 @xcite .
note that the essential difference of the many - body electronic structure for the lattice constants of gaas and inas is completely due to the difference in the position of the fermi energy with respect to the minority - spin band gap whereas the self - energy characterizing the correlation effects is not changed too much ( fig .
[ sigma ] ) .
the total density of states @xmath62 is rather sensitive to the difference between the band edge @xmath63 and the fermi energy @xmath64 .
if this difference is very small ( i.e. , the system is close to the electronic topological transition @xmath65 ) one can use a simple expression for the singular contribution to the bare density of states , @xmath66 @xmath67 .
the appearance of the complex self - energy @xmath68 changes the singular contribution as @xmath69^{1/2}\ ] ] where @xmath70 ( cf .
ref . ) . assuming that the self - energy is small in comparison with @xmath71 one can find for the states in the gap @xmath72 @xmath73 one can see that the shift of the gap edge changes drastically the density of states for the same @xmath74 in the remaining part of the paper we elaborate more on the possibility of the practical use of tunable properties of non - quasiparticle states in cras grown on different substrates . for
most of the applications room - temperature and the stability of the ferromagnetic state are the most important prerequisites .
the ferromagnetic cras material might be grown on iii - v semiconductors similarly to the zinc - blende crsb @xcite .
we observed the presence of the nqp states for cras lattice parameters larger than @xmath75 .
it was found experimentally that at 300@xmath76 , around this value of the lattice parameter a stable solid solution of ga@xmath77in@xmath78as is formed @xcite .
so , from the practical point of view a @xmath79 of gallium in a ga@xmath80in@xmath81as compound would constitute the ideal substrate on which the cras half - metal would present tunable properties of its nqps .
having the ga@xmath77in@xmath78as substrate cras could be a part of the epitaxial iii - v structure providing an easy way to integrate with the existing semiconductor technology .
in addition , to determine possible substrates for growth of layered half - metallic materials , electronic structure calculations were carried out for lattice parameters in the range @xmath82 to @xmath83 @xcite . according to these calculations @xcite , growth with minimal strain
might be accomplished in a half - metallic multilayer system grown on inas substrate which would be the best choice to evidence the nqp states , since the fermi energy is situated in this case far enough from the bottom of the conduction band .
a very high sensitivity of the minority - electron density of states near the fermi energy to the lattice constant opens a new interesting opportunity to run the current in cras - based tunnel junction .
suppose we have an antiparallel orientation of the magnetizations in the tunnel junction ( such as shown in fig .
[ model]c ) , then the @xmath84 characteristic is determined by the density of the non - quasiparticle states .
thus if we will influence the lattice constant ( e.g. , using a piezoelectric material ) one can modify the differential conductivity .
this makes cras a very promising material with tunable characteristics which opens new ways for applications in spintronics .
this work is part of the research programme of the stichting voor fundamenteel onderzoek der materie ( fom ) , which is financially supported by the nederlandse organisatie voor wetenschappelijk onderzoek ( nwo ) . v. yu . irkhin , m. i. katsnelson , and a. i. lichtenstein , cond - mat/0406487 ; to appear in proceedings of the workshop `` local - moment ferromagnets : unique properties for modern applications '' ( berlin , march 2004 ) .
.the muffin - tin radii of the cr , as and the two types of empty spheres placed as vacancies of the zb structure .
the constreined lda values for the coulomb and exchange interactions are presented . [ cols="^,^,^,^,^,^,^ " , ] | correlation effects on the electronic structure of half - metallic cras in zinc - blende structure are studied for different substrate lattice constants . depending on the substrate the spectral weight of the non - quasiparticle states
might be tuned from a well developed value in the case of inas substrate to an almost negligible contribution for the gaas one . a piezoelectric material that would allow the change in the substrate lattice parameters opens the possibility for practical investigations of the switchable ( tunable ) non - quasiparticle states . since the latter are important for the tunneling magnetoresistance and related phenomena it creates new opportunities in spintronics .
15.6 cm |
in view of its importance in explaining the observational cosmology many authors have considered cosmological models with dark energy . in a recent paper kremer @xcite has modelled the universe as a binary mixture whose constitutes are described by a van der waals fluid and by a dark energy density .
@xcite showed that `` tracker field '' , a form of qiuntessence , may explain the coincidence , adding new motivation for the quintessence scenario .
the fate of density perturbations in a universe dominated by the chaplygin gas , which exhibit negative pressure was studied by fabris _
_ @xcite . model with chaplygin gas was also studied in the refs .
@xcite . in doing
so the author considered a spatially flat , homogeneous and isotropic universe described by a friedmann - robertson - walker ( frw ) metric .
since the theoretical arguments and recent experimental data support the existence of an anisotropic phase that approaches an isotropic one , it makes sense to consider the models of universe with anisotropic back - ground in presence of dark energy .
the simplest of anisotropic models , which nevertheless rather completely describe the anisotropic effects , are bianchi type - i ( bi ) homogeneous models whose spatial sections are flat but the expansion or contraction rate is direction - dependent . in a number of papers ,
e.g. , @xcite , we have studied the role of a @xmath2 term in the evolution of a bi space - time in presence of spinor and/or scalar field with a perfect fluid satisfying the equation of state @xmath0 . in this paper we study the evolution of an initially anisotropic universe given by a bi spacetime in presence of a perfect fluid obeying not only @xmath0 , but also the van der waals equation of state .
the einstein field equation on account of the cosmological constant we write in the form @xmath3 here @xmath4 is the ricci tensor , @xmath5 is the ricci scalar and @xmath6 is the einstein gravitational constant . as was mentioned earlier , @xmath2 is the cosmological constant . to allow a steady state cosmological solution to the gravitational field equations einstein @xcite introduced a fundamental constant , known as cosmological constant or @xmath2 term , into the system .
soon after e. hubble had experimentally established that the universe is expanding , einstein returned to the original form of his equations citing his temporary modification of them as the biggest blunder of his life .
@xmath2 term made a temporary comeback in the late 60 s .
finally after the pioneer paper by a. guth @xcite on inflationary cosmology researchers began to study the models with @xmath2 term with growing interest .
note that in our previous papers @xcite we studied the einstein field equations where the cosmological term appears with a negative sign . here
following the original paper by einstein and one by sahni @xcite we choose the sign to be positive . in this paper
a positive @xmath2 corresponds to the universal repulsive force , while a negative one gives an additional gravitational force .
note that a positive @xmath2 is often taken to a form of dark energy .
we study the gravitational field given by an anisotropic bianchi type i ( bi ) cosmological model and choose it in the form : @xmath7 with the metric functions @xmath8 being the functions of time @xmath9 only .
the einstein field equations for the bi space - time in presence of the @xmath2 term now we write in the form [ ee ] @xmath10 here over - dot means differentiation with respect to @xmath9 .
the energy - momentum tensor of the source is given by @xmath11 where @xmath12 is the flow vector satisfying @xmath13 here @xmath14 is the total energy density of a perfect fluid and/or dark energy density , while @xmath15 is the corresponding pressure . @xmath15 and @xmath14 are related by an equation of state which will be studied below in detail . in a co - moving system of
coordinates from one finds @xmath16 in view of from one immediately obtains @xcite [ abc ] @xmath17 , \label{a } \\
b(t ) & = & d_{2 } \tau^{1/3 } \exp \bigl[x_2 \int\,\frac{dt}{\tau ( t ) } \bigr ] , \label{b}\\ c(t ) & = & d_{3 } \tau^{1/3}\exp \bigl[x_3 \int\,\frac{dt}{\tau ( t ) } \bigr ] .
\label{c}\end{aligned}\ ] ] here @xmath18 and @xmath19 are some arbitrary constants obeying @xmath20 and @xmath21 is a function of @xmath9 defined to be @xmath22 from for @xmath21 one find @xmath23 on the other hand the conservation law for the energy - momentum tensor gives @xmath24 after a little manipulations from and we find @xmath25 with @xmath26 being an arbitrary constant .
let us now , in analogy with hubble constant , define @xmath27 on account of from one derives @xmath28 it should be noted that the energy density of the universe is a positive quantity .
it is believed that at the early stage of evolution when the volume scale @xmath21 was close to zero , the energy density of the universe was infinitely large . on the other hand with
the expansion of the universe , i.e. , with the increase of @xmath21 , the energy density @xmath14 decreases and an infinitely large @xmath21 corresponds to a @xmath14 close to zero .
say at some stage of evolution @xmath14 is too small to be ignored
. in that case from follows @xmath29 as it is seen from in this case @xmath2 is essentially non - negative .
we can also conclude from that in absence of a @xmath2 term beginning from some value of @xmath21 the evolution of the universe comes stand - still , i.e. , @xmath21 becomes constant , since @xmath30 becomes trivial , whereas in case of a positive @xmath2 the process of evolution of the universe never comes to a halt .
moreover it is believed that the presence of the dark energy ( which can be explained with a positive @xmath2 as well ) results in the accelerated expansion of the universe .
as far as negative @xmath2 is concerned , its presence imposes some restriction on @xmath14 , namely , @xmath14 can never be small enough to be ignored .
it means in that case there exists some upper limit for @xmath21 as well ( note that @xmath21 is essentially nonnegative , i.e. bound from below ) . in our previous papers we came to the same conclusion @xcite [ with a positive @xmath2 which in the present paper appears to be negative ] .
inserting and into one now finds @xmath31 in view of from follows that if the perfect fluid is given by a stiff matter where @xmath32 , the corresponding solution does not depend on the constant @xmath33 .
let us now go back to the eq . .
it is in fact the first integral of and can be written as @xmath34 on the other hand , rewriting in the form @xmath35 and taking into account that @xmath15 is a function of @xmath14 , one concludes that the right hand side of the eq . is a function of @xmath21 only , i.e. , @xmath36 from a mechanical point of view eq . can be interpreted as an equation of motion of a single particle with unit mass under the force @xmath37 . then the following first integral exists @xcite : @xmath38}\ , .
\label{1stint}\ ] ] here @xmath39 can be viewed as energy and @xmath40 is the potential of the force @xmath41 . comparing the eqs .
and one finds @xmath42 and @xmath43 let us finally write the solution to the eq . in quadrature : @xmath44 where the integration constant @xmath45 can be taken to be zero , since it only gives a shift in time . in what follows we study the eqs . and
for perfect fluid obeying different equations of state .
in this section we consider the case when the source field is given by a perfect fluid . here
we study two possibilities : ( i ) the energy density and the pressure of the perfect fluid are connected by a linear equation of state ; ( ii ) the equation of state is a nonlinear ( van der waals ) one . in this subsection
we consider the case when the source field is given by a perfect fluid fluid obeying the equation of state @xmath47 where @xmath48 is a constant and lies in the interval @xmath49 $ ] . depending on its numerical value
, @xmath48 describes the following types of universes @xcite [ zeta ] @xmath50 in view of , from for the energy density and pressure one obtains @xmath51 where @xmath52 is the constant of integration . for @xmath21 from one finds @xmath53 as one sees , the positivity of the radical in for a negative @xmath2 imposes some restriction on the upper value of @xmath21 , i.e. , @xmath21 should be bound from above as well . in fig .
[ potential ] the graphical view of the potential @xmath54 is illustrated for a negative @xmath2 . as it was mentioned earlier , @xmath39 or @xmath33 in case of @xmath55
does not play any role .
universe in this case initially expands , reaches to the maximum and then begin to contract finally giving rise to a space - time singularity [ cf fig .
[ oscillation ] ] .
for the other cases depending on the choice of @xmath39 expansion of the universe is either non - periodic [ fig .
[ nonperiodic ] ] with a singularity at the end or oscillatory one without space - time singularity [ fig .
[ oscillation ] ] . in fig .
[ inflation ] we demonstrate the evolution of the bi universe with a positive @xmath2 . in this case
the universe expands exponentially , the initial anisotropy quickly dies away and the bi universe evolves into a isotropic frw one .
there does not any upper bound for @xmath21 in case of a positive @xmath2 .
note that in the figs .
( [ potential ] - [ inflation ] ) @xmath56 , @xmath57 , @xmath58 and @xmath59 stand for dust , radiation , hard universe and stiff matter , respectively .
1 cm in absence of the @xmath2 term one immediately finds @xmath60 with @xmath61 being some integration constant .
as one sees from , in absence of a @xmath2 term , for @xmath62 the initially anisotropic universe eventually evolves into an isotropic frw one , whereas , for @xmath55 , i.e. , in case of stiff matter the isotropization does not take place .
here we consider the case when the source field is given by a perfect fluid with a van der waals equation of state in absence of dissipative process .
the pressure of the van der waals fluid @xmath63 is related to its energy density @xmath64 by @xcite @xmath65 in the pressure and the energy density is written in terms of dimensionless reduced variables and @xmath66 is a parameter connected with a reduced temperature . in the figs .
[ enprlm1 ] and [ enprlp1 ] the energy density and the pressure of the system are illustrated with a negative and a positive @xmath2 term as well as in absence of it .
inserting into on account of one finds @xmath67}{2(3 \kappa - \{3 h^2 - \lambda - c_1/(3 \tau^2)\})}\nonumber \\ & + & \frac{3}{2\kappa}\bigl(\{3 h^2 - \lambda - c_1/(3 \tau^2)\}\bigr)^2 . \label{hwes}\end{aligned}\ ] ] it can be easily verified that the eq . in absence of @xmath68 term and @xmath69 and @xmath70
coincides with that given in @xcite : @xmath71 . \label{hwes1}\ ] ] the solution of the second - order differential equation for @xmath72 can be found by specifying the initial value for @xmath72 at @xmath73 , for a given value of parameter @xmath66 .
here we graphically present some results concerning the evolution of bi universe with a van der waals fluid . in fig .
[ infl1 ] we compare the evolution of @xmath21 with and without @xmath2 term . as one sees , the character of evolution does not depend on the sign of @xmath2 . in all cases we find exponentially expanding universe , though the rapidity of growth depends on @xmath2 . the fig .
[ volume ] gives the comparison of the expansion of @xmath21 with perfect fluid obeying different equations of state .
the evolution of an anisotropic universe given by a bianchi type i cosmological model is studied in presence of a perfect fluid and a @xmath2 term .
it has been shown that in case of a perfect fluid obeying @xmath0 , where @xmath15 and @xmath14 are the pressure and energy density of the fluid , respectively , a negative @xmath2 may generate an oscillation in the system thus giving rise to a singularity - free mode of expansion .
introduction of a positive @xmath2 in this case results in a rapid expansion of the universe .
if the universe is filled with a van der waals fluid , no oscillatory or non - periodic mode of expansion occurs .
independent to the sign of @xmath2 the universe in this case expands exponentially .
a. einstein , sitzungsber .
. wiss . * 1 * , 142 ( 1917 ) and * 1 * , 349 ( 1919 ) [ english translation : _ the principle of relativity _ ( methuen , 1923 , reprinted by dover , new york , 1924 ) , pp . 177 and 191 ] . | we consider a self - consistent system of bianchi type - i ( bi ) gravitational field and a binary mixture of perfect fluid and dark energy given by a cosmological constant .
the perfect fluid is chosen to be the one obeying either the usual equation of state , i.e. , @xmath0 , with @xmath1 $ ] or a van der waals equation of state .
role of the @xmath2 term in the evolution of the bi universe has been studied . |
theories that extend the standard model of particle physics have presented scenarios that allow for , or even predict , spatial - temporal variations of the constants of nature @xcite .
possible variations of the fine structure constant , @xmath1 , representing the strength of the electromagnetic force , or the proton - to - electron mass ratio , @xmath2 , a measure of the strength of the strong force , lead to shifts in the spectra of atoms and molecules .
many studies have been devoted to observe these shifts . by comparing metal absorptions in the spectra from distant quasars with the corresponding transitions measured in the laboratory , webb _ et al . _
@xcite found evidence that suggests that the fine structure constant , @xmath1 , has a smaller value at high redshift .
in later work , this variation was interpreted as a spatial variation of @xmath1 @xcite . in parallel , laboratory experiments on earth
are used to probe possible variations in the current epoch .
compared to their astrophysical counterpart , their advantage is their great accuracy , reproducibility and unequivocal interpretation . by comparing transitions in different isotopes of dysprosium ,
a possible variation of the fine structure constant was found to be @xmath3/yr @xcite . whereas atomic spectra are mostly sensitive to variations in @xmath1 , molecular spectra
can be used to detect a possible variation of @xmath2 .
the most stringent independent test of the time variation of @xmath2 in the current epoch is set by comparing vibrational transitions in with a cesium fountain , which has resulted in a limit for the variation of @xmath4 of 5.6@xmath510@xmath6/yr @xcite .
tests of @xmath2-variation on cosmological time scales have been performed by comparing spectra of molecular hydrogen measured in the laboratory with those observed at redshifts @xmath7 , corresponding to a look - back time of @xmath8gyr , constraining @xmath9 @xcite .
the most stringent limit on a variation of @xmath2 in the early universe are set by bagdonaite _
@xcite from comparing absorptions by methanol in objects at a redshift of 0.89 , corresponding to a look - back time of 7gyr , with laboratory data . the uncertainty in the constraint derived by bagdonaite _
is dominated by the precision of the astrophysical data . however , when more accurate astrophysical data become available , the error in the laboratory data will become significant . in this paper
, we investigate the possibilities to increase the precision of selected microwave transitions in methanol .
we focus on the four transitions in observed by bagdonaite _
et al . _ , and two transitions in that provided that the precision is significantly enhanced
might be used for a laboratory test of the time variation of @xmath2 .
line centers of methanol transitions in the microwave region are typically obtained from absorption measurements in a gas cell , resulting in ( doppler - limited ) measurement uncertainties around 50khz corresponding to a relative uncertainty of @xmath1010@xmath11 @xcite .
for a limited number of lines higher resolution data was obtained by a pulsed molecular beam fabry - perot fourier - transform microwave spectrometer of the balle - flygare type @xcite , reaching accuracies around 20khz @xcite . using a beam - maser setup ,
two single methanol transitions were recorded with relative accuracies of @xmath1010@xmath0 @xcite .
so far , this is the only study that was able to ( partly ) resolve hyper - fine structure in methanol .
all these studies are based on detecting absorption or emission of the microwave field .
a significantly higher precision seems feasible in a rabi - type setup using lasers to state - selectively detect the methanol molecules .
unfortunately , so far no suitable state - selective detection scheme for methanol has been demonstrated .
the only study that reports the detection of methanol by resonance - enhanced multi - photon ionization ( rempi ) , involved either the repulsive @xmath12 rydberg state or one of the @xmath13 rydberg state ; both resulting in broad unresolved bands @xcite . here
, we explore the possibility for detecting methanol molecules using a femtosecond laser , while relying on inhomogeneous electric fields to separate the different quantum states present in the beam .
this paper is organized as follows : in sec .
[ sec : methanol ] we discuss the energy level structure of methanol , and review the origin of the large sensitivity coefficients that are found in this molecule .
furthermore , we outline the procedure that was adopted to calculate the stark interaction for methanol . in sec . [
sec : simulations ] we simulate trajectories of methanol molecules through a typical beam resonance setup , using the derived stark shifts as input . in sec .
[ sec : detection ] , we present measurements that determine the efficiency of ionizing methanol molecules using femtosecond laser pulses . finally , in sec .
[ sec : allandev ] , we use the simulations and measured ion yield to estimate the expected accuracy of the described beam resonance setup .
methanol is the simplest representative of the class of alcohol molecules and consists of a hydroxyl ( ) group attached to a methyl group ( ) . the bond that connects the two parts of the molecule is flexible , allowing the methyl group to rotate with respect to the hydroxyl group .
this rotation is hindered by a threefold potential barrier with minima and maxima that correspond to a staggered and eclipsed configuration of the two groups , respectively .
for the lowest energy levels , the internal rotation or torsion is classically forbidden and only occurs due to quantum mechanical tunneling of the hydrogen atoms . in order to account for this additional degree of freedom ,
the normal asymmetric top hamiltonian has to be augmented with a term that describe the torsion motion . to simplify the calculation ,
the coupling between overall and internal rotation is partly eliminated by applying an axis transformation to the coordinates of the hamiltonian ( the so - called rho - axis method " or ram ) . in the rho - axis method ,
the full torsion - rotation hamiltonian for methanol becomes @xcite : @xmath14 where @xmath15 , @xmath16 , @xmath17 and @xmath18 represent the torsion , overall rotation , centrifugal distortion , and higher - order torsion - rotation interaction terms , respectively .
this hamiltonian is implemented in the belgi code @xcite that we have used to calculate the level energies of methanol .
the current version of the code was modified and improved by xu _
et al . _
@xcite in a number of ways useful for treating the large datasets available for the methanol molecule .
furthermore , the code has been optimized to make it faster and a substantial number of higher order parameters has been added .
using a set of 119 molecular constants for from ref .
@xcite and 54 constants for from ref .
@xcite , the lower energy levels are found with an accuracy @xmath19khz .
the hamiltonian of eq .
is diagonalized in a two - step process to obtain the torsion - rotation energy levels @xcite . in the first step ,
the torsional hamiltonian ( @xmath15 ) is diagonalized : @xmath20 where @xmath21 is the internal rotation constant , @xmath22 is the ratio of the moment of inertia of the methyl top relative to the moments of inertia of the molecule as a whole , and @xmath23 is the internal rotation potential barrier , @xmath24 is the internal rotation angular momentum , @xmath25 is the projection of the global rotation on the @xmath26 molecular axis and @xmath1 is the torsional angle .
the eigenvalues obtained after this first step are the torsional energies for each torsional level that are characterized by the quantum numbers @xmath27 , @xmath28 and @xmath29 ( a species ) or @xmath30 ( e species ) .
the @xmath31 and @xmath32 symmetry species can be considered as two different molecular species in the same sense as ortho- and para ammonia .
the torsional eigenfunctions can be written as linear combinations of the basis wave functions @xcite : @xmath33\alpha\right ) } } , \label{eq : torseigenfunc}\ ] ] where @xmath34 is an integer . in the second step , the rotational factor @xmath35 in the eigenfunctions from eq .
is replaced by the full symmetric top wave function @xmath36 to generate the basis set used to diagonalize the remaining terms of the hamiltonian from eq . , i.e. @xmath16 , @xmath17 , and @xmath18 . the lowest energy levels of and are calculated at zero electric field using the molecular constants from refs . @xcite , and are shown on the left and right - hand side of fig . [
fig : k - ladders ] , respectively .
the arrangement of energy levels within a symmetry state resembles that of a prolate symmetric top , with the difference being that every @xmath27 ladder obtains an additional energy offset due to the @xmath27 dependent tunneling splitting . as a consequence , certain states in neighboring
@xmath27 ladders may become nearly degenerate .
it was shown by jansen _
et al . _
@xcite and levshakov _ et al .
_ @xcite that transitions between these nearly degenerate states are very sensitive to possible variations of the proton - to - electron mass ratio , @xmath2 .
the sensitivity of a transition with a frequency @xmath37 is defined as @xmath38 the overall rotational energy of the molecule scales with its rotational constants and is thus inversely proportional to the reduced mass of the system .
therefore , pure rotational transitions have a sensitivity coefficient , @xmath39 .
the torsional energy arises from the tunneling effect and similar to the inversion splitting in ammonia
depends exponentially on the effective mass that tunnels . for the normal isotopologue of methanol ,
the sensitivity coefficient of a purely torsional transition has a value of @xmath40 .
note that , due to symmetry , such transitions are not allowed in methanol .
an interesting effect occurs for transitions between different @xmath27 ladders . in this case ,
part of the overall rotational energy is converted into internal rotation energy or vice versa . when the energies involved are rather similar
i.e. , when the levels are nearly degenerate this results in enhanced sensitivity coefficients .
these enhancements occur generally in every internal rotor molecule , but because of a number of favorable properties , the effect is exceptionally large in methanol @xcite . the arrows in the panel on the left - hand side of fig .
[ fig : k - ladders ] mark transitions that have been observed in the study of bagdonaite _ et al . _
the two transitions in the @xmath41 ladder are pure rotational transitions and have a sensitivity coefficient of @xmath39 .
the transitions between the @xmath41 and @xmath42 ladder have sensitivities of @xmath43 and @xmath44 .
the shaded area in the panel on the right - hand side of fig . [ fig : k - ladders ] highlights the near degeneracies present between the @xmath45 , @xmath46 , and @xmath47 levels in .
the @xmath48 transition and the @xmath49 have sensitivity coefficients of 330 and -42 , respectively .
thus , if @xmath2 increases , the frequency of the @xmath48 transition becomes larger while the frequency of the @xmath49 transition becomes smaller . by comparing these two transitions over a number of years , a possible variation of @xmath2 can be constrained or measured . in order to calculate the stark shift on the energy levels of methanol for different values of the electric field ,
we have included the stark hamiltonian @xmath50 in the second diagonalization step of the belgi code .
the stark term is given by @xmath51 where @xmath52 is the body - fixed electric dipole moment vector and @xmath53 the electric field . in our calculation , @xmath52 is represented by the body - fixed dipole moments @xmath54 and @xmath55 along the @xmath56 and @xmath57 axes of the ram frame , respectively . for , @xmath58d and @xmath59d @xcite , while for , @xmath60d and @xmath61d @xcite .
the matrix elements of @xmath50 are taken from eqs .
( 1)(4 ) of kleiner _
et al . _
since the stark effect induces nonzero matrix elements @xmath62 , the hamiltonian matrix was extended to include those interactions . for a given @xmath63 value
, the hamiltonian matrix only has off - diagonal blocks involving the nearby @xmath64 states .
each @xmath65 energy level is split into @xmath66 components , characterized by @xmath67 , the projection of the total angular momentum @xmath63 along the laboratory axis @xmath68 .
@xmath67 is the only good quantum number in the presence of the electric field , hence the basis used for calculating the stark effect for a certain @xmath67 level includes all states with this specific @xmath67 .
our approach was tested by comparing the stark shifts calculated by the modified version of belgi with the results of a perturbation - like calculation using a code written by lees and coauthors @xcite . for @xmath63 levels up to @xmath69 ,
the ratios of the stark shifts obtained by these two methods are close to unity at 10kv / cm and range from 0.7 to 1 at 50kv / cm . in the left and right - hand side of fig .
[ fig : stark_curves ] , energies of the near - degenerate @xmath70 and @xmath45 levels of and @xmath45 , @xmath46 , and @xmath47 levels of , respectively , are plotted as function of the applied electric - field strength .
the lower panels of fig .
[ fig : stark_curves ] show the corresponding effective dipole moments @xmath71 , defined as @xmath72 where @xmath73 is the stark shift of the quantum state in an electric field of magnitude @xmath74 .
the effective dipole moment is a measure for the orientation of the molecule in a specific state , i.e. , the expectation value of the dipole moment in a space - fixed axes system .
note that in the hamiltonian used , @xmath75 levels are degenerate with @xmath76 levels ( if @xmath77 and @xmath78 ) and , as these levels have different parity , this degeneracy results in a effective dipole moment that is non - zero at 0kv / cm .
this non - physical result disappears when high - order couplings are incorporated @xcite . in table
[ tab : mu_eff ] the effective dipole moments of a selection of levels are listed .
1 @ p-0.5 cm d .. 0 d .. 2 d .. 2 c d .. 3 d .. 3 d .. 11 d .. 7 + & & & & & + + + & & & & sym & & & & + + + + + & 0 & 0 & 0 & @xmath79 & 127.683 & 127.475 & 6.5410 ^ -3 & 1.0310 ^ -1 + & 2 & -1 & -2 & @xmath32 & 136.400 & 135.995 & 8.3410 ^ -3 & 2.0810 ^ -2 + & 2 & -1 & -1 & @xmath32 & 136.400 & 136.140 & 6.0910 ^ -3 & 2.0810 ^ -2 + & 2 & -1 & 0 & @xmath32 & 136.400 & 136.292 & 3.6510 ^ -3 & 2.0810 ^ -2 + & 2 & -1 & 1 & @xmath32 & 136.400 & 136.455 & 4.1410 ^ -4 & 2.0810 ^ -2 + & 2 & -1 & 2 & @xmath32 & 136.400 & 136.586 & -1.1010 ^ -3 & 2.0810 ^ -2 + & 0 & 0 & 0
& @xmath32 & 136.805 & 136.701 & 3.0010 ^ -3 & 3.7010 ^ -2 + & 1 & 0 & -1 & @xmath32 & 138.419 & 138.425 & -3.4510 ^ -5 & 2.3310 ^ -2 + & 1 & 0 & 0 & @xmath32 & 138.419 & 138.575 & -4.6610 ^ -3 & 2.3310 ^ -2 + & 1 & 0 & 1 & @xmath32 & 138.419 & 138.427 & -1.9410 ^ -4 & 2.3310 ^ -2 + & 1 & 1 & -1 & @xmath79 & 139.388 & 138.932 & 8.1110 ^ -3 & 1.7710 ^ -3 + & 1 & 1 & 0 & @xmath79 & 139.388 & 139.392 & -1.2510 ^ -4 & 1.7710 ^ -3 + & 1 & 1 & 1 & @xmath79 & 139.388 & 138.932 & 8.1110 ^
-3 & 1.7710 ^ -3 + & 3 & -1 & -3 & @xmath32 & 141.240 & 140.908 & 7.2010 ^ -3 & 5.1610 ^ -3 + & 3 & -1 & -2 & @xmath32 & 141.240 & 140.899 & 7.6510 ^ -3 & 5.1610 ^ -3 + & 3 & -1 & -1 & @xmath32 & 141.240 & 140.927 & 7.1810 ^ -3 & 5.1610 ^ -3 + & 3 & -1 & 0 & @xmath32 & 141.240 & 140.978 & 6.3210 ^ -3 & 5.1610 ^ -3 + & 3 & -1 & 1 & @xmath32 & 141.240 & 141.051 & 5.2810 ^ -3 & 5.1610 ^ -3 + & 3 & -1 & 2 & @xmath32 & 141.240 & 141.154 & 3.8710 ^ -3 & 5.1610 ^ -3 + & 3 & -1 & 3 & @xmath32 & 141.240 & 141.354 & 1.5210 ^ -5 & 5.1610 ^ -3 + & 2 & 0 & -2 & @xmath32 & 141.646 & 141.752 & -2.2810 ^ -3 & 9.1910 ^ -3 + & 2 & 0 & -1 & @xmath32 & 141.646 & 141.845 & -4.5710 ^ -3 & 9.1910 ^ -3 + & 2 & 0 & 0 & @xmath32 & 141.646 & 141.884 & -5.5310 ^ -3 & 9.1910 ^ -3 + & 2 & 0 & 1 & @xmath32 & 141.646 & 141.872 & -5.2110 ^ -3 & 9.1910 ^ -3 + & 2 & 0 & 2 & @xmath32 & 141.646 & 141.801 & -3.6810 ^ -3 & 9.1910 ^ -3 + + + + & 2 & 0 & -2 & @xmath32 & 133.587 & 133.557 & 1.4310 ^ -3 & 9.9610 ^ -3 + & 2 & 0 & -1 & @xmath32 & 133.587 & 133.362 & 4.7610 ^ -3 & 9.9610
^ -3 + & 2 & 0 & 0 & @xmath32 & 133.587 & 132.989 & 1.0510 ^ -2 & 9.9610 ^ -3 + & 2 & 0 & 1 & @xmath32 & 133.587 & 132.766 & 1.3210 ^ -2 & 9.9610 ^ -3 + & 2 & 0 & 2 & @xmath32 & 133.587 & 132.961 & 1.3410 ^ -2 & 9.9610 ^ -3 + & 1 & 1 & -1 & @xmath32 & 133.635 & 133.730 & -2.3410 ^ -3 & 4.9110 ^ -3 + & 1 & 1 & 0 & @xmath32 & 133.635 & 133.713 & -3.0310 ^ -3 & 4.9110 ^ -3 + & 1 & 1 & 1 & @xmath32 & 133.635 & 133.586 & 1.0210 ^ -3 & 4.9110 ^ -3 + & 2 & 2 & -2 & @xmath32 & 133.675 & 134.156 & -6.1310 ^ -3 & 4.8610 ^ -3 + & 2 & 2 & -1 & @xmath32 & 133.675 & 134.600 & -1.6010 ^ -2 & 4.8610 ^ -3 + & 2 & 2 & 0 & @xmath32 & 133.675 & 134.266 & -9.6410 ^ -3 & 4.8610 ^ -3 + & 2 & 2 & 1 & @xmath32 & 133.675 & 133.837 & -3.2410 ^ -3 & 4.8610 ^ -3 + & 2 & 2 & 2 & @xmath32 & 133.675 & 133.557 & 1.0510 ^ -3 & 4.8610
in this section , we use the obtained effective dipole moments to simulate molecular trajectories through a typical beam resonance setup @xcite . we consider a molecular beam apparatus that consists of ( i ) a collimation section ; ( ii ) an electric deflection field for state preparation ; ( iii ) a microwave cavity ; ( iv ) a second electric deflector for state - selection and ( v ) a detector . in the calculations , we assume a methanol beam with a mean velocity of 800 m/s .
this beam is collimated by two 0.6 mm diameter skimmers that are separated by 500 mm .
the deflection fields used for pre and post state selection consist of a cylindrical electrode and a parabolically shaped electrode to which a voltage difference is applied .
the electric field inside such a deflector was analyzed by de nijs and bethlem in terms of a multipole expansion @xcite .
it was shown that the most optimal deflection field is created by choosing a field that only contains a dipole and quadrupole term , the strength of which is represented by @xmath80 and @xmath81 , respectively . in our calculations , we take @xmath80=18kv and @xmath81=3.6kv while all other expansion terms are set to 0 . furthermore , @xmath82 that characterizes the size of the electrodes is taken as 3 mm ( i.e. , the distance between the electrodes is @xmath83 mm ) .
this results in an electric field magnitude of 60 kv / cm on the molecular beam axis . for the deflectors are assumed to be 200 mm long , while for a length of 150 mm is used .
both deflectors are oriented in the same direction , i.e. , a molecule that is deflected upwards in the first electric field , will again be deflected upwards in the second electric field , if it has not undergone a microwave transition .
the force on a molecule inside one of the deflection fields is found from : @xmath84 where the gradient of the electric field is calculated from an analytical expression derived in de nijs and bethlem @xcite .
the two deflection fields are separated by 200 mm .
after the last deflection field , the molecules travel 200 mm further before being ionized in the focus of a femtosecond laser .
the total flight path adds up to 1350 mm for and 1250 mm for .
the 200 mm free flight between the deflection fields can be used to drive a microwave transition . in our simulations ,
the trajectories of typically @xmath85 molecules are calculated .
the position and velocity spread are sampled randomly from a gaussian distribution .
the initial quantum state is sampled from a boltzmann distribution that assumes a temperature of either 5 or 10k . as mentioned earlier ,
@xmath31 and @xmath32 type methanol should be considered as two different molecular species . in @xmath31-type methanol ,
the total spin is @xmath86 , whereas in @xmath32-type methanol @xmath87 , hence , the nuclear spin degeneracy for @xmath31-type methanol is twice that of @xmath32-type methanol .
however , this is matched by the presence of both @xmath88 and @xmath89 levels of @xmath32 symmetry @xcite . in our calculations , we assume equal numbers of molecules of @xmath31 and @xmath32 species for , whereas the @xmath90 ratio in is taken to be 11/16 . is assumed to be preserved in the field - free region , whereas for the blue curve complete @xmath67 scrambling is assumed .
the curves are given an offset for clarity .
[ fig : mueff ] ] fig .
[ fig : mueff ] shows the distribution of the effective dipole moment of the molecules .
the black curve ( a ) shows the distribution when the deflection fields are off and the laser focus is situated at the molecular beam axis .
the red ( b ) curve shows this distribution when the deflection fields are on and the laser focus is situated 2.2 mm above the molecular beam axis .
it is observed that only molecules are detected in states that have an effective dipole moment of @xmath91 10@xmath92@xmath93/(kv / cm ) . with these settings ,
the resolution of the selector , @xmath94 , is 0.2@xmath5 10@xmath92@xmath93/(kv / cm ) . in these simulations , the @xmath67 state of the molecule
is assumed to be preserved
i.e. , it is assumed that a homogeneous electric field is applied between the deflectors to keep the molecules oriented
. however , in our experiment the region where the microwave excitation takes place should be completely shielded from external magnetic or electric fields . in this case
the different @xmath95 levels are degenerate and , in the worst case scenario , the @xmath95 distribution after the interaction zone is randomized completely .
the results of a simulation that assumes complete de - orientation is shown as the blue curve ( c ) in fig .
[ fig : mueff ] .
clearly , @xmath95 scrambling greatly reduces the effectiveness of the state selection . note that due to hyperfine splittings
, the degeneracy of the different levels at zero electric field is lifted .
the hyperfine structure has not been taken into account in our simulations as it is largely unknown , in fact , resolving the hyperfine splittings is an important motivation for this study .
axis when the microwave field is off resonance ( black curve ) and on resonance with the @xmath96 transition ( red curve ) , assuming a rotational temperature of 5k .
both curves are normalized to the number of molecules that are detected at @xmath97 when the deflection fields are turned off . _
lower panel : _ difference between the count rate on and off resonance .
[ fig : deflection ] ] fig .
[ fig : deflection ] shows a simulation of what would be observed if the vertical ( @xmath98 ) position of the laser focus is scanned , i.e. , the spatial distribution of the methanol beam in the direction of field - gradient .
the red curve shows the distribution that is observed when the microwave field drives 50% of the molecules from the @xmath99 to @xmath100 level , and _
vice versa_. we assume that the microwave field only drives @xmath101 transitions .
the black curve shows the distribution when no micro - wave field is present . in both cases the @xmath95 distribution
after the interaction zone is assumed to be randomized completely . both curves are normalized to the number of molecules that are detected at @xmath97 when the deflection fields are turned off .
we define the count rate with the microwave field being on or off resonance as @xmath102 and @xmath103 , respectively .
the count rate of the undeflected beam is defined as @xmath104 . in order to observe if a transition has occurred , we look for a difference between @xmath103 and @xmath102 , i.e. , the difference between the black and red curves in fig . [
fig : deflection ] .
this difference is shown in the lower panel of fig .
[ fig : deflection ] . if we choose the vertical position of the laser focus to be 2.3 mm above the molecular beam axis , @xmath105=
0.062 and @xmath106=0.048 .
this results in a normalized averaged count rate , @xmath107 , and a difference in count rate , @xmath108 . in sec .
[ sec : allandev ] , it is shown that the accuracy scales with @xmath109 . for the @xmath110 transition , this number is equal to 0.056 .
table [ tab : transitions ] lists @xmath111 and @xmath109 for the selected transitions assuming a rotational temperature of 5 and 10k .
the errors in @xmath111 and @xmath109 are estimated to be @xmath112 for and @xmath113 for . for the selected transitions in @xmath109 is @xmath114 at 5k and slightly less at 10k .
for the two considered transitions in , these numbers are even less favorable , which is slightly surprising given that the effective dipole moments of the levels involved are rather different ; the different @xmath95 states of the @xmath115 level all have a positive stark shift , the @xmath95 states of the @xmath100 level have a negative stark effect , whereas the @xmath95 states of the @xmath116 level have virtually no stark shift .
however , the fact that the effective dipoles of the different @xmath95 states within the levels varies considerably , combined with @xmath95 scrambling in the interrogation zone , complicates state - selection .
furthermore , the population of the levels involved is rather small .
1 @l d .. 3 d .. 2 d .. 3 d .. 2 d .. 4 d .. 2 d .. 2 d .. 4 + transition , @xmath117 & & & & + + + & & & & & & & & + + + + + @xmath118 & 12178.587 & -33 & 2.3 & 0.013 & 0.056 & 2.5 & 0.017 & 0.063 + @xmath119 & 48372.460 & -1.00 & -3.5 & 0.028 & 0.078 & -3.7 & 0.012 & 0.045 + @xmath120 & 48376.887 & -1.00 & 2.1 & 0.008 & 0.028 & -1.7 & 0.012 & 0.026 + @xmath121 & 60531.477 & -7.4 & 1.6 & 0.012 & 0.041 & -1.2 & -0.016 & 0.029 + + + + @xmath122 & 1202.296 & 330 & 2.1 & -0.003 & 0.018 & -4.2 & -0.002 & 0.017 + @xmath123 & 1424.219 & -42 & -1.7 & 0.018 & 0.033 & -3.4 & 0.004 & 0.023 + +
in order to estimate the efficiency of non - resonant multi - photon ionization of methanol by a femtosecond laser , we have performed test measurements in an existing molecular beam machine described elsewhere @xcite .
a supersonic molecular beam is prepared by expanding a mixture of @xmath100.5% in argon through a 200@xmath2 m diameter pulsed piezo nozzle @xcite operating at a repetition frequency of 101000hz . the pressure behind the nozzle was kept below 1 bar in order to prevent cluster formation .
the molecular beam passes a 1.5 mm and a 1.0 mm skimmer , before entering the detection region where it is intersected at right angle with a focused ( @xmath124 mm ) femtosecond laser beam .
the total flight path from the nozzle to the detection zone adds up to 160 mm .
the produced ions are accelerated towards a position sensitive microchannel - plate detector mounted in front of a fast phosphor screen where they are counted .
the femtosecond pulses are generated by a commercial amplified regen laser system ( spectra physics spitfire ) that produces a 1khz pulse train and is tunable around 800 nm with an output energy of about 800@xmath2j and a duration of 120 fs .
the fundamental 800 nm light is doubled in a bbo crystal to obtain a 400 nm laser pulse with an energy of about 100@xmath2j . as the ionization potential of methanol is 10.84ev @xcite , at least four photons of 3.10ev
are required to ionize it .
table [ tab : ioncount ] lists the number of detected parent ions per second using an intensity of 20 or 100@xmath2j per pulse at a 1khz repetition rate .
although the ion yield can probably be increased by further optimalization of the molecular beam parameters and focal properties of the laser beam , it seems unlikely that this changes the obtained results by more than a factor of 2 .
the fact that the nonresonant multi - photon ionization rate does not follow a @xmath125 dependence , with @xmath126 the laser intensity and @xmath127 the number of photons absorbed , can be understood by considering the intensity profile of the laser focus . at a certain laser intensity , @xmath128 , the ionization rate at the beam waist reaches unity , and the signal increases due to the increase in volume for which @xmath129 . for comparison ,
the ion yield for fluoromethane is also listed in table [ tab : ioncount ] .
note that the ionization potential of fluoromethane is 12.50ev @xcite and at least five photons are required to ionize it
. 1 @l c c c + molecule & ie(ev ) & + + + & & 20@xmath2j / pulse & 100@xmath2j / pulse + + + & 10.84 & 1200 & 20000 + & 12.50 & 3500 & 25000 + +
from the measured ion yield and the results from the simulations , we can now estimate the accuracy that can be obtained in the described beam machine .
in general , the accuracy of a frequency measurement depends on the @xmath130 factor of the system and the signal to noise ratio , @xmath131 . in an experiment that relies on counting individual ions
the accuracy is expressed by the allan deviation @xcite @xmath132 where @xmath133 defines the duration of one measurement cycle , @xmath134 is the total measurement time , @xmath37 is the frequency of the measured transition and @xmath135 the width of the spectral line . in our simulations , we assume that the microwave cavity ( or microwave cavities , if we use a ramsey type setup ) has a length of 160 mm while the molecular beam has a velocity of 800 m/s .
this implies that the total interrogation time for the measurement is 200@xmath136 , corresponding to a spectral width of @xmath104khz . for the @xmath70 to 2@xmath137 transition at 12ghz ,
this results in a q - factor of @xmath138 .
the @xmath131 depends on the number of ions that are detected per cycle .
if we spend half the duration of a measurement cycle on and off resonance , the number of ions that contribute to the signal is given by @xmath139 .
assuming poissonian statistics , the noise on the total number of detected ions is given by @xmath140 .
thus the allan deviation becomes ; @xmath141 as expected , the allan deviation becomes infinite when the count rates on and off resonance are equal , while it scales with @xmath142 when @xmath102 is zero ( background free ) .
the extra factor of @xmath143 arises because we spend half of the time on signal and half of the time off signal .
the values for @xmath109 are given in table [ tab : transitions ] .
@xmath104 , the count rate when the deflection fields are turned off , can be estimated from the test measurements .
assuming that the density in the molecular beam decreases quadratically with the distance from the nozzle , the expected signal at a distance of 1350 mm behind the nozzle is about a factor of 64 smaller than that obtained in the test setup , hence @xmath144 ions / sec .
for the @xmath70 to 2@xmath137 transition at 12ghz , the allan deviation is @xmath145 . in order to achieve a fractional accuracy of 10@xmath0 a measurement time of about 80 minutes
would be required .
similar results are found for the other transition in listed in table [ tab : transitions ] .
the expected accuracy for is less favorable .
in this paper we estimate the feasibility of performing high resolution microwave spectroscopy on selected transitions in methanol using a rabi - type molecular beam setup in combination with a femtosecond laser . we have adapted the belgi programme to calculate the stark effect for the different isotopologues of methanol .
the calculated stark shifts are reasonably large , typically between -1 and 1@xmath93 in a field of 100kv / cm for the different rotational states .
thus , the molecules can be easily manipulated using modest sized deflection fields . due to the small rotational constants of methanol ,
many states are populated even at the low temperatures that can be obtained in supersonic beams . with the resolution obtained in a typical molecular beam deflection setup ,
it is not possible to select individual quantum states .
typically , 5 or more states are present within the laser focus .
furthermore , in the field free region used for inducing the microwave transition , the @xmath95 quantum number is not preserved , leading to a reduction of the state purity . note that , @xmath67-scrambling is both a blessing and a cure , on one hand it reduces the observed difference between @xmath102 and @xmath103 , on the other hand it ensures that all hyperfine levels are addressed without he need to change the position of the laser focus .
@xmath67-scrambling can be enforced by rapidly switching the second deflection field on entrance of the molecules @xcite . from simulations ,
we find that the differences in signal on or off resonance , @xmath111 , are in the range of @xmath146 for the selected transitions in , and slightly less for the two considered transitions in . in order to estimate the detection efficiency of methanol molecules using a femtosecond laser , we have performed test measurements in an existing molecular beam machine .
using a laser power of 100@xmath2j per pulse at 1khz repetition rate , @xmath147 ions per second were detected at a distance of 160 mm behind the nozzle . from these numbers , the described molecular beam deflection setup seems suited to measure microwave transitions in at an accuracy below 10@xmath0 , although the required measurement times are rather long .
currently the laser system is being upgraded to provide peak intensities that are 5 times higher than used in this study .
this should result in an increase of the ion yield of at least a factor of 25 , making this experiment perfectly feasible . due to the small differences in count rate
, however , it might be tedious to find the optimal position of the laser focus . for a laboratory test using an accuracy well below 10@xmath148 would be required .
this seems unlikely to be achieved in a beam machine as considered here .
this research has been supported by fom via a projectruimte grant .
r.m.l . acknowledges financial support from nserc canada . w.u .
acknowledges financial support from the templeton foundation .
h.l.b acknowledges financial support from nwo via a vidi grant and from the erc via a starting grant .
we thank c.s .
lehmann for helpful discussions and assistance during the experiment .
lias , ionization energy evaluation , in _ nist chemistry webbook , nist standard reference database number 69 _ , edited by p. j. linstrom and w. g. mallard ( national institute of standards and technology , gaithersburg md , 20899 , ( retrieved january 31 , 2013 ) ; http://webbook.nist.gov ) , http://webbook.nist.gov . | recently , the extremely sensitive torsion - rotation transitions in methanol have been used to set a tight constraint on a possible variation of the proton - to - electron mass ratio over cosmological time scales . in order to improve this constraint ,
laboratory data of increased accuracy will be required . here
, we explore the possibility for performing high - resolution spectroscopy on methanol in a stark - deflected molecular beam .
we have calculated the stark shift of the lower rotational levels in the ground torsion - vibrational state of and molecules , and have used this to simulate trajectories through a typical molecular beam resonance setup .
furthermore , we have determined the efficiency of non - resonant multi - photon ionization of methanol molecules using a femtosecond laser pulse .
the described setup is in principle suited to measure microwave transitions in at an accuracy below 10@xmath0 . |
we acknowledge the financial support of the royal society ( s.b.d . ) and the uk epsrc , and invaluable discussions with igor mazin , michelle johannes and zahid hasan .
this experiment was performed with the approval of the japan synchrotron radiation research institute ( jasri , proposal nos .
2005a0092-nd3a - np and 2005b0182-nd3a - np ) .
this work was partially supported by a grant - in - aid for scientific research ( no.18340111 ) from the ministry of education , culture , sports , science and technology , japan .
k. takada , h. sakurai , e. takayama - muromachi , f. izumi , r. a. dilanian , t. sasaki , nature ( london ) * 422 * , 53 ( 2003 ) .
q. h. wang , d. h. lee , p. a. lee , phys .
b * 69 * , 092504 ( 2004 ) .
i. terasaki , y. sasago , k. uchinokura , phys .
b * 56 * , r12685 ( 1997 ) .
y. y. wang , n. s. rogado , r. j. cava , n. p. ong , nature ( london ) * 423 * , 425 ( 2003 ) .
m. d. johannes , i. i. mazin , d. j. singh , d. a. papaconstantopoulos , phys .
* 93 * , 097005 ( 2004 ) .
k. kuroki , y. tanaka , r. arita , phys .
* 93 * , 077001 ( 2004 ) .
i. i. mazin , m. d. johannes , nature physics * 1 * , 91 ( 2005 ) .
m. mochizuki , y. yanase , m. ogata , phys .
lett . * 94 * , 147005 ( 2005 ) .
m. mochizuki and m. ogata , j. phys .
japan * 75 * , 113703 ( 2006 ) .
m. z. hasan , y. d. chuang , d. qian , y. w. li , y. kong , a. p. kuprin , a. v. fedorov , r. kimmerling , e. rotenberg , k. rossnagel , z. hussain , h. koh , n. s. rogado , m. l. foo , r. j. cava , phys .
lett . * 92 * , 246402 ( 2004 ) .
h. b. yang , s. c. wang , a. k. p. sekharan , h. matsui , s. souma , t. sato , t. takahashi , t. takeuchi , j. c. campuzano , r. jin , b. c. sales , d. mandrus , z. wang , h. ding , phys .
lett . * 92 * , 246403 ( 2004 ) .
h. b. yang , z. h. pan , a. k. p. sekharan , t. sato , s. souma , t. takahashi , r. jin , b. c. sales , d. mandrus , a. v. fedorov , z. wang , h. ding , phys .
lett . * 95 * , 146401 ( 2005 ) .
d. qian , l. wray , d. hsieh , d. wu , j. l. luo , n. l. wang , a. kuprin , a. fedorov , r. j. cava , l. viciu , m. z. hasan , phys .
lett . * 96 * , 046407 ( 2006 ) .
d. qian , d. hsieh , l. wray , y. d. chuang , a. fedorov , d. wu , j. l. luo , n. l. wang , m. viciu , r. j. cava , m. z. hasan , phys .
lett . * 96 * , 216405 ( 2006 ) .
t. shimojima , k. ishizaka , s. tsuda , t. kiss , t. yokoya , a. chainani , s. shin , p. badica , k. yamada and k. togano , phys .
lett . * 97 * , 267003 ( 2006 ) .
l. viciu , j. w. g. bos , h. w. zandbergen , q. huang , m. l. foo , s. ishiwata , a. p. ramirez , m. lee , n. p. ong , r. j. cava , phys .
b * 73 * , 174104 ( 2006 ) . d. j. singh , phys . rev .
b * 61 * , 13397 ( 2000 ) .
p. h. zhang , w. d. luo , m. l. cohen , s. g. louie , phys .
93 * , 236402 ( 2004 ) .
s. zhou , m. gao , h. ding , p. a. lee , z. q. wang , phys .
lett . * 94 * , 206401 ( 2005 ) .
q. huang , m. l. foo , r. a. pascal , j. w. lynn , b. h. toby , t. he , h. w. zandbergen , r. j. cava , phys .
b * 70 * , 184110 ( 2004 ) .
j. wooldridge , d. m. paul , g. balakrishnan , m. r. lees , j. phys .
matter * 17 * , 707 ( 2005 ) .
f. c. chou , j. h. cho , y. s. lee , phys . rev .
b * 70 * , 144526 ( 2004 ) .
f. c. chou , j. h. cho , p. a. lee , e. t. abel , k. matan , y. s. lee , phys .
lett . * 92 * , 157004 ( 2004 ) .
n. hiraoka , m. itou , t. ohata , m. mizumaki , y. sakurai , n. sakai , j. synch .
8 * , 26 ( 2001 ) .
y. sakurai , m. itou , j. phys .
solids * 65 * , 2061 ( 2004 ) .
g. kontrym - sznajd , phys .
status solidi a * 117 * , 227 ( 1990 ) .
major , s. b. dugdale , r. j. watts , g. santi , m. a. alam , s. m. hayden , j. a. duffy , j. w. taylor , t. jarlborg , e. bruno , d. benea , h. ebert , phys .
lett . * 92 * , 107003 ( 2004 ) .
s. b. dugdale , r. j. watts , j. laverock , zs . major , m. a. alam , m. samsel - czekaa , g. kontrym - sznajd , y. sakurai , m. itou , d. fort , phys .
* 96 * , 046406 ( 2006 ) .
m. d. johannes , d. j. singh , phys .
b * 70 * , 014507 ( 2004 ) .
r. j. xiao , h. x. yang , j. q. li , phys .
b * 73 * , 092517 ( 2006 ) . l. balicas , j. g. analytis , y. j. jo , k. storr , h. zandbergen , y. xin , n. e. hussey , f. c. chou , p. a. lee , phys .
97 * , 126401 ( 2006 ) .
j. p. rueff , m. calandra , m. dastuto , ph .
leininger , a. shukla , a. bossak , m. krisch , h. ishii , y. cai , p. badica , t. sasaki , k. yamada and k. togano , phys .
b * 74 * , 020504(r ) ( 2006 ) m. mochizuki , y. yanase , m. ogata , j. phys .
. jpn . * 74 * , 1670 ( 2005 ) .
d. j. singh , d. kasinathan , phys .
rev . lett . * 97 * 016404 ( 2006 ) .
n. oeschler , r. a. fisher , n. e. phillips , j. e. gordon , m. l. foo , r. j. cava , chinese journal of physics * 43 * , 574 ( 2005 ) . k. w. lee , w. e. pickett , phys . rev .
b * 72 * , 115110 ( 2005 )
. h. ishida , m. d. johannes , a. liebsch , phys .
lett . * 94 * , 196401 ( 2005 ) .
h. ishida and a. liebsch , cond - mat/0705.3627 a. bourgeois , a. a. aligia , t. kroll , m. d. nunez - regueiro , phys .
b * 75 * , 174518 ( 2007 ) .
b. kumar , b. s. shastry , phys .
b * 68 * , 104508 ( 2003 ) .
m. ogata , j. phys .
. jpn . * 72 * , 1839 ( 2003 ) .
a. tanaka , x. hu , phys . rev
. lett . * 91 * , 257006 ( 2003 ) .
g. baskaran , phys .
* 91 * , 097003 ( 2003 ) . | the surprise discovery of superconductivity below 5k in sodium cobalt oxides when hydrated with water has caught the attention of experimentalists and theorists alike .
most explanations for its occurence have focused heavily on the properties of some small elliptically shaped pockets predicted to be the electronically dominant fermi surface sheet , but direct attempts to look for them have instead cast serious doubts over their existence .
here we present evidence that these pockets do indeed exist , based on bulk measurements of the electron momentum distribution in unhydrated and hydrated sodium cobalt oxides using the technique of x - ray compton scattering .
hydrated sodium cobalt oxides ( na@[email protected]@xmath2o ) , for a certain range of na concentrations , exhibit superconductivity at a temperature of 5k @xcite .
although many analogies have been drawn with the high-@xmath3 cuprates ( for instance , as being possibly the only other example of a mott insulator becoming superconducting under doping @xcite ) , the sodium cobaltate system exhibits its own unique set of anomalous behaviour such as unusually high thermopower @xcite and @xmath4-linear resistivity @xcite ) , distinctly indicative of strongly correlated electron behaviour . in a conventional superconductor ,
electrons at the fermi surface form cooper pairs under an attractive interaction mediated by lattice vibrations .
the manner in which electrons form these pairs can be strongly influenced by the shape of the fermi surface .
questions regarding the origin of the pairing interaction and the nature of the superconductivity in the cobaltates has stimulated significant theoretical speculation , most of which has focused heavily on the properties of some small elliptically shaped pockets predicted to be the electronically dominant fermi surface sheet @xcite , but the outcome of direct attempts to look for them has instead cast serious doubt over their existence @xcite . here
we present evidence that these pockets do indeed exist , based on bulk measurements of the electron momentum distribution in unhydrated and hydrated sodium cobalt oxides using the technique of x - ray compton scattering .
the structure of na@xmath0coo@xmath2 comprises hexagonal planes of electronically active edge - sharing coo@xmath5 octahedra @xcite .
these planes are separated by insulating layers of na , and , in the hydrated samples , water , that serve as spacers ( resulting in electronic two - dimensionality ) and charge reservoirs .
the intercalation of water , at a concentration of @xmath6 , has a dramatic effect on properties of the compound .
it is accompanied by a near doubling of the @xmath7-axis lattice parameter and , for @xmath8 , the onset of superconductivity .
calculations of electronic structure based on the local - density - approximation ( lda ) predict , for concentrations @xmath90.6 in na@xmath0coo@xmath2 , a fermi surface composed of two hole sheets originating from co @xmath10-orbitals @xcite .
the larger sheet , of @xmath11 character , is centred at the @xmath12-point ( zone centre ) whereas the other sheet , of @xmath13 character , comprises six smaller elliptical cylinders located between @xmath12 and @xmath14 ( for @xmath15 , this band becomes completely filled and does not contribute to the fermi surface ) .
these @xmath13 pockets may be of considerable importance , since many of the models for superconductivity in na@xmath0coo@xmath16h@xmath2o are predicated upon their existence .
in particular , an analysis of permitted order - parameter symmetries has indicated that an unconventional @xmath17-wave ( @xmath18 ) would be the most probable if the pockets exist @xcite .
with 70% of the density - of - states at the fermi level , these pockets are central to several models of spin - fluctuation - mediated superconductivity , developing either via imperfect nesting between the fermi surface pockets @xcite , or as a consequence of these disconnected pieces of fermi surface @xcite . a series of angle - resolved photoemission spectroscopy ( arpes ) results
, the consensus of which report observation of the @xmath11 sheet at @xmath12 but crucially not the @xmath13 pockets , has fuelled the controversy surrounding the elliptical pockets .
indeed , they observe the @xmath13 band as lying completely below the fermi level @xcite .
however , it is well - known that surface effects and matrix elements can have a strong influence on photoemission results , although there are recent reports of measurements being made with an electron escape depth of 200 @xcite . here
we directly tackle this controversy by presenting a compton scattering study both of the fermi surfaces of a representative set of compositions of the unhydrated parent compound na@xmath0coo@xmath2 ( @xmath19=0.38 , 0.51 and 0.74 ) and a hydrated ( actually deuterated ) sample at a superconducting composition , na@xmath20coo@xmath21d@xmath2o .
for these measurements , single crystals of @xmath22 were grown in warwick using the floating - zone technique .
samples of lower sodium concentration were then obtained by a chemical deintercalation method , immersing the crystals in solutions of br@xmath23 and acetonitrile .
the lattice parameters were obtained by x - ray diffraction ; the relationship between na doping and the crystal structure is well characterised by powder neutron diffraction measurements and icp - aes techniques @xcite and so the sodium concentrations for the three crystals used in this study were determined as 0.74(1 ) , 0.51(1 ) and 0.38(1 ) .
the macroscopic properties ( magnetic susceptibility , heat capacity and transport measurements @xcite ) are identical to those previously reported for similar compositions @xcite .
the superconducting sample was produced by the chemical intercalation of deuterium oxide ( d@xmath2o was used rather than h@xmath2o to allow future neutron experiments on the same sample ) by submersion in liquid d@xmath2o for three months at 5@xmath24c @xcite .
subsequent to the compton experiment , a measurement of its magnetisation showed that it was superconducting at a temperature of 3.5k .
a compton profile represents a double integral ( one - dimensional projection ) of the full three - dimensional electron momentum density . for each composition , five compton profiles equally spaced between @xmath12-@xmath25 and @xmath12-@xmath14 were measured on the high - resolution compton spectrometer of beamline bl08w at the spring-8 synchrotron .
the unhydrated measurements were made at room temperature , while those on the hydrated sample were at 11k .
the hydrated sample was transferred from a d@xmath2o bath directly onto a cryostat precooled to @xmath26k under a helium atmosphere in order to preserve the hydration . on removal at the end of the measurement , the sample
was observed to be still in its hydrated state .
the spectrometer consists of a cauchois - type crystal analyser and a position - sensitive detector , with a resolution fwhm at the compton peak of 0.115 a.u ( 1 a.u . of momentum @xmath27 1.99 @xmath2810@xmath29 kg m s@xmath30 ) @xcite .
for each compton profile , @xmath31 counts in the peak data channel were accumulated , and each compton profile was corrected for possible multiple - scattering contributions .
a two - dimensional momentum density , representing a projection down the @xmath7-axis of the full three - dimensional density , was reconstructed from each set of five profiles using tomographic techniques @xcite and then folded back into the first bz using the lock - crisp - west procedure @xcite to obtain the occupation density from which the occupied parts of the bz could be inferred .
the occupation density is shown for each composition in fig .
[ expfs ] , where black represents the lowest occupancy and white the highest .
( color online ) the experimental fermi surface of na@xmath0coo@xmath2 for ( a ) @xmath32 , ( b ) 0.51 , and ( c ) 0.38 , and for na@xmath20coo@xmath21d@xmath2o ( d ) obtained from the reconstruction of five compton profiles for each composition .
the boundary of the first brillouin zone is indicated . ] considering first the unhydrated parent compound , the contours associated with the hexagonal @xmath11 hole sheet can be clearly identified for @xmath32 , but the hexagonal shape becomes progressively less clear and is significantly distorted in the @xmath33 data .
we shall argue that this distortion is strong evidence for the presence of the @xmath13 pockets .
fermi surface obtained by plotting the contour at the maximum of the first derivative of the occupation density for na@xmath34coo@xmath2 . ] to assess the size of the hexagonal fermi surface of the na@xmath34coo@xmath2 compound , a method using the extrema in the first derivative of the occupation density was employed ( see for example @xcite ) .
at this composition elliptical pockets are not expected , and so the determination using this method should unambiguously reveal the hexagonal fermi surface .
fig .
[ fderiv ] is the result , and shows a fermi surface in excellent agreement with lda calculations .
we can explain the distortion of the hexagonal shape for smaller na concentrations as being due to the presence of @xmath13 elliptical hole pockets close to the central @xmath11 fermi surface .
a simple geometric simulation of such a fermi surface is shown in fig .
[ simul ] together with the resulting occupation density ( convoluted with the experimental resolution ) , illustrating how the presence of small pockets distort the hexagonal appearance of the @xmath11 fermi surface . the experimental occupation density for the hydrated na@xmath20coo@xmath21d@xmath2o ( fig .
[ expfs ] ) is also strongly suggestive of the presence of small hole pockets , which can be discerned close to the central hexagon ; it is also worth remarking that the @xmath11 sheet retains a strong hexagonal shape , whereas electronic structure calculations in the hydrated structure ( although without the actual presence of water ) predict something more circular @xcite .
our results suggest that the effects of hydration on the fermi surface are in fact rather modest , and perhaps not as drastic as suggested by xiao _
et al .
_ @xcite .
an estimate of the areas of the hexagonal @xmath11 sheet and ( where appropriate ) the six elliptical @xmath13 pockets based on a comparison of simulations to the experimental data ( with the total area constrained by the appropriate na concentration ) is presented in table [ fsparam ] . that the pockets consistently appear rather close to the @xmath11 sheet is also noteworthy .
( color online ) a simulation of a fermi surface comprising a central hexagonal sheet ( representing the @xmath11 hole sheet ) and six @xmath13 elliptical hole pockets ( left ) together with the resulting occupation density convoluted with the experimental resolution ( right ) .
, title="fig : " ] ( color online ) a simulation of a fermi surface comprising a central hexagonal sheet ( representing the @xmath11 hole sheet ) and six @xmath13 elliptical hole pockets ( left ) together with the resulting occupation density convoluted with the experimental resolution ( right ) . , title="fig : " ] .
estimate of areas of the @xmath11 and e@xmath35 fermi surfaces ( in the case of the latter , this is the total area of all six pockets ) based on applying the simulation of the fermi surface as illustrated in fig . [ simul ] for the compositions studied , as a proportion of the hexagonal first brillouin zone . [ cols="^,^,^",options="header " , ] recent measurement of shubnikov - de haas oscillations in na@xmath36coo@xmath2 indicate the presence of some unidentified fermi surface pockets occupying approximately 0.6% and 1.4% of the bz @xcite , which is consistent with our estimate of the size of each pocket we observe occupying about 0.8% ( table [ fsparam ] ) .
in addition , an examination of phonon softening in this system by rueff _
et al . _
@xcite is interpreted by those authors as strong evidence for the existence of nested pockets .
the question of why the arpes experiments have consistently not observed these e@xmath35 pockets remains .
issues such as surface sensitivity , including possible surface relaxations of coo@xmath5 octahedral contractions that destroy the pockets @xcite , as well as matrix - element effects or na disorder @xcite must be possibilities .
however , at least in the superconducting compound , it is very difficult to reconcile the observed behaviour of the specific heat , or even understand the presence of superconductivity without the presence of the @xmath13 pockets @xcite .
moreover , in an attempt to take into account coulomb correlations not included in the lda , calculations @xcite based on the lda@xmath37 approach , have suggested that for a sufficiently large coulomb energy ( @xmath38 ev ) the @xmath13 band is pulled below the fermi level , eliminating these smaller fermi surface pockets for all na concentrations .
however , other studies have put an upper limit of about 2.3ev on @xmath39 @xcite , and when dynamical coulomb correlations are incorporated the effect is to stabilize the @xmath13 pockets @xcite .
the theoretical debate rages on , with predictions in support of @xcite and contrary to @xcite the existence of these pockets . in conclusion
, we have presented the fermi surface of several members ( @xmath33 , 0.51 and 0.74 ) of the unhydrated sodium cobalt oxide na@xmath0coo@xmath2 and of a hydrated composition na@xmath20coo@xmath21d@xmath2o .
reasonable _ qualitative _ agreement is observed between our experimentally determined fermi surfaces and the lda predictions , and there is clear evidence for the smaller e@xmath35 elliptical hole pockets which develop at lower na concentrations than na@xmath34coo@xmath2 . for na@xmath40coo@xmath2 , their presence
is clearly indicated in experimental maps of the occupancy within the brillouin zone .
most importantly , however , the occupancy map for na@xmath20coo@xmath21d@xmath2o also shows the presence of small e@xmath35 elliptical hole pockets . while alternative models that describe the superconductivity arising as a consequence of frustration on the triangular lattice @xcite , nesting across the large , hexagonal sheet @xcite , or spin fluctuations enhanced by the dopant dynamics @xcite are not ( and can not be ) ruled out , the observation of the pockets lends strong support to theories based on their special nesting properties |
as is well known , the event horizon of black holes hides the initial conditions of the collapsed matter other than mass , charge , and angular momenta , which is named the black hole no - hair theorem @xcite .
the similar thing should happen for inflation because of the cosmological event horizon , namely , any initial conditions should go away beyond the cosmological event horizon .
in fact , in the presence of the cosmological constant , there is a cosmic no - hair theorem proved by wald @xcite . even for a general accelerating universe driven by a scalar field ,
it is legitimate to expect that the anisotropy decays exponentially first in the presence of the cosmological event horizon .
this prejudice is often referred to as the cosmic no - hair conjecture .
historically , there have been challenges to the cosmic no - hair conjecture @xcite .
unfortunately , it turned out that these models suffer from either the instability , or a fine tuning problem , or a naturalness problem @xcite . recently , however , we have succeeded in finding stable anisotropic inflationary solutions in the context of supergravity .
more precisely , we have shown that , in the presence of a gauge field coupled with an inflaton , there could be small anisotropy in the expansion rate which never decays during inflation .
since anisotropic inflation is an attractor , this can be regarded as a counter example to the cosmic no - hair conjecture @xcite . moreover , primordial fluctuations created quantum mechanically during inflation also exhibit statistical anisotropy @xcite .
indeed , from the point of view of precision cosmology , it is important to explore the role of gauge fields in inflation @xcite .
it is well known that the supergravity models can be constrained by comparing predictions of inflation with cosmological observations .
for example , the tilt of the spectrum gives interesting information of the superpotential @xmath0 and the kaler potential @xmath1 which are functionals of complex scalar fields . here , a bar represents a complex conjugate .
more concretely , the bosonic part of the action of supergravity reads @xmath2\end{aligned}\ ] ] where we have defined kaler metric @xmath3 and a covariant derivative @xmath4 . here , @xmath5 represents the reduced planck mass .
we note that the gauge kinetic function @xmath6 in front of the gauge kinetic term @xmath7 could be nontrivial functions of the scalar fields .
curiously , so far , the gauge field part has been neglected in most of discussion of inflation .
the reason is partially due to the cosmic no - hair conjecture which states that the anisotropy , curvature , and any matter will be diluted once inflation commences .
however , as we will see soon , this is not true when we look at percent level fine structures of inflationary scenarios .
we find that gauge fields play an important role in the early universe .
in fact , we will show that there exists anisotropic inflation and the degree of the anisotropy we found is smaller than the slow roll parameter .
apparently , the expansion is almost isotropic .
nevertheless , anisotropic inflation provides a qualitatively new picture of the universe . in this paper , as a demonstration , we consider the cases where the functional form of the potential and the gauge kinetic functions has an exponential dependence on the inflaton
. in particular , we present exact anisotropic inflationary solutions .
moreover , we argue anisotropic inflation becomes an attractor in the phase space .
the organization of the paper is as follows . in section 2 ,
we introduce inflationary models where the gauge field couples with an inflaton and obtain exact solutions which contain anisotropic power - law inflationary solutions . in section 3 , we show that the anisotropic inflation is an attractor in the phase space .
thus , generic trajectories converge to anisotropic inflation .
this implies that the cosmic no - hair conjecture does not hold in general .
the final section is devoted to the conclusion .
in this section , we consider a simple model with exponential potential and gauge kinetic functions and then find exact power - law inflationary solutions . in addition to a well known isotropic power - law solution , we find an anisotropic power - law inflationary solution .
we consider the following action for the metric @xmath8 , the inflaton field @xmath9 and the gauge field @xmath10 coupled with @xmath9 : @xmath11 \ , \label{action1}\end{aligned}\ ] ] where @xmath12 is the determinant of the metric , @xmath13 is the ricci scalar , respectively .
the field strength of the gauge field is defined by @xmath14 . motivated by the dimensional reduction of higher dimensional theory such as string theory
, we assume the exponential potential and the exponential gauge kinetic functions @xmath15 in principle , the parameters @xmath16 , and @xmath17 can be determined once the compactification scheme is specified .
however , we leave those free hereafter . using the gauge invariance
, we can choose the gauge @xmath18 . without loosing the generality
, we can take the @xmath19-axis in the direction of the gauge field .
hence , we have the homogeneous fields of the form @xmath20 and @xmath21 as there exists the rotational symmetry in the @xmath22-@xmath23 plane , we take the metric to be @xmath24 \ , \label{metric}\end{aligned}\ ] ] where the cosmic time @xmath25 is used . here
, @xmath26 is an isotropic scale factor and @xmath27 represents a deviation from the isotropy .
it is easy to solve the equation for the gauge field as @xmath28 where @xmath29 denotes a constant of integration .
using eq.([eq : ax ] ) , we obtain equations @xmath30 \ , \label{hamiltonian:0}\\ \ddot{\alpha } & = & -3\dot{\alpha}^2 + \frac{1}{m_p^2 } v(\phi ) + \frac{p_{a}^2}{6m_p^2}f^{-2}(\phi ) e^{-4\alpha -4\sigma } , \label{evolution : alpha:0}\\ \ddot{\sigma } & = & -3\dot{\alpha}\dot{\sigma } + \frac{p_{a}^2}{3m_p^2}f^{-2}(\phi ) e^{-4\alpha -4\sigma } \label{eq : sigma:0 } , \\ \ddot{\phi } & = & -3\dot{\alpha}\dot{\phi } -v'(\phi ) + p_{a}^2 f^{-3}(\phi ) f'(\phi ) e^{-4\alpha -4\sigma } \label{eq : phi:0 } \ .\end{aligned}\ ] ] here , an overdot and a prime denote the derivative with respect to the cosmic time @xmath25 and @xmath9 , respectively . in the absence of gauge fields
, it is known that there exists the power - law inflationary solution .
therefore , let us first seek the power - law solutions by assuming @xmath31 apparently , for a trivial gauge field @xmath32 , we have the isotropic power - law solution @xmath33 in this case , we have the spacetime @xmath34 hence , we need @xmath35 for obtaining the accelerating expansion . next , interestingly , we see that there exists the other non - trivial exact solution in spite of the existence of the no - hair theorem @xcite . from the hamiltonian constraint equation ( [ hamiltonian:0 ] )
, we set two relations @xmath36 to have the same time dependence for each term . the latter relation is necessary only in the non - trivial gauge field case , @xmath37 .
then , for the amplitudes to balance , we need @xmath38 where we have defined @xmath39 the equation for the scale factor ( [ evolution : alpha:0 ] ) under eq .
( [ a ] ) yields @xmath40 similarly , the equation for the anisotropy ( [ eq : sigma:0 ] ) gives @xmath41 finally , from the equation for the scalar ( [ eq : phi:0 ] ) , we obtain @xmath42 using eqs .
( [ a ] ) , ( [ c ] ) and ( [ d ] ) , we can solve @xmath43 and @xmath44 as @xmath45 substituting these results into eq .
( [ e ] ) , we obtain @xmath46 = 0 \ .\end{aligned}\ ] ] in the case of @xmath47 , we have @xmath48 .
hence , it is not our desired solution .
thus , we have to choose @xmath49 substituting this result into eq .
( [ c ] ) , we obtain @xmath50 from eq .
( [ a ] ) , we have @xmath51 finally , eq . ( [ uw ] ) reduce to @xmath52 note that eq .
( [ b ] ) is automatically satisfied . in order to have inflation , we need @xmath35 . for these cases
, @xmath43 is always positive .
since @xmath44 should be also positive , we have the condition @xmath53 hence , @xmath54 must be much larger than one .
now , the spacetime reads @xmath55 this describes the anisotropically inflating universe .
the average expansion rate is determined by @xmath56 and the average slow roll parameter is given by @xmath57 where we have defined @xmath58 . in the limit @xmath35 and @xmath59 , this reduces to @xmath60 .
now , the anisotropy is characterized by the ratio between the anisotropic expansion rate and the isotropic expansion rate @xmath61 remarkably , the anisotropy is proportional to the slow roll parameter and the parameter @xmath62 is smaller than 1 .
hence , typically , the anisotropy is quite small .
although the anisotropy is always small , it persists during inflation .
therefore , its effect on the cosmological observables is not negligible .
now , we know two exact solutions , isotropic and anisotropic power - law inflationary solutions . these solutions are not general solutions but particular solutions . in order to eamine if the cosmic no - hair conjecture holds or not , we need to know the fate of generic solutions . to this end , the dynamical analysis is useful .
since two exact solutions correspond to two fixed points in the phase space , we did the stability analysis around the fixed points @xcite .
it turns out that the isotropic fixed point indicated by orange circle in fig.[fg : phase ] is a saddle point which is an attractor only on the two - dimensional @xmath63 plane .
thus , in the absence of gauge fileds , isotropic inflation is an attractor .
namely , the cosmic no - hair holds in this special case . on the other hand
, we found that the anisotropic fixed point is stable .
thus , the end point of trajectories around the unstable isotropic power - law inflation must be anisotropic power - law inflation . in fig.[fg : phase ] , we depicted the phase flow in the phase space for @xmath64 . we see that the trajectories converge to the anisotropic fixed point indicated by yellow circle .
when the initial conditions are close to the @xmath63 plane , the trajectory approaches to the saddle point , but eventually it goes into the attractor .
interestingly , even if we start from isotropic universe , the final state of the universe is anisotropic ( green line ) .
this can be regarded as the spontaneous breakdown of the rotational symmetry . .
the axis @xmath65 represents the energy density of the gauge field , the axis @xmath66 describes the anisotropy of the universe , and the axis @xmath67 denotes the velocity of the inflaton .
the yellow and orange circles indicate the anisotropic and isotropic fixed points respectively .
the trajectories converge to the anisotropic fixed point.,width=472 ] as we have shown , clearly , anisotropic inflation gives rise to a counter example to the cosmic no - hair conjecture .
this feature is not specific to the exponential potential .
although , in general cases , we do not have exact solutions , qualitative feature is almost the same . indeed , there are many concrete examples which violate the cosmic no - hair conjecture @xcite .
we should note that the cosmological constant is assumed in the cosmic no - hair theorem proved by wald @xcite . in the case of inflation
, the inflaton can mimic the cosmological constant .
hence , the comic no - hair holds in the conventional cases .
however , in the presence of a non - trivial coupling between the inflaton and the gauge field , the cosmic no - hair theorem can not be applicable anymore @xcite .
we have examined a supergravity model with a gauge kinetic function .
it turned out that exact anisotropic power - law inflationary solutions exist when both the potential function for an inflaton and the gauge kinetic function are exponential type .
we showed that the degree of the anisotropy is proportional to the slow roll parameter .
the slow roll parameter depends both on the potential function and the gauge kinetic function .
the dynamical system analysis told us that the anisotropic power - law inflation is an attractor .
therefore , the result we have mentioned in this paper presents a clear counter example to the cosmic no - hair conjecture .
the imprint of anisotropic inflation can be found in the cosmic microwave background radiation @xcite .
we should note that the statistical anisotropy could be large even when the anisotropy in the expansion is quite small .
the result in this paper also has implication in cosmological magnetic fields @xcite .
i would like to thank sugumi kanno and masa - aki watanabe for collaborations .
this work was supported in part by the grants - in - aid for scientific research ( c ) no.25400251 and grants - in - aid for scientific research on innovative areas no.26104708 .
99 l. h. ford , phys . rev .
d * 40 * , 967 ( 1989 ) ; n. kaloper , phys .
d * 44 * , 2380 ( 1991 ) ; s. kawai and j. soda , phys .
d * 59 * , 063506 ( 1999 ) ; j. d. barrow and s. hervik , phys .
d * 73 * , 023007 ( 2006 ) ; l. ackerman , s. m. carroll and m. b. wise , phys .
d * 75 * , 083502 ( 2007 ) [ erratum - ibid .
d * 80 * , 069901 ( 2009 ) ] ; a. golovnev , v. mukhanov and v. vanchurin , jcap * 0806 * , 009 ( 2008 ) ; s. kanno , m. kimura , j. soda and s. yokoyama , jcap * 0808 * , 034 ( 2008 ) .
b. himmetoglu , c. r. contaldi and m. peloso , phys . rev
. lett . * 102 * , 111301 ( 2009 ) ; a. golovnev , phys .
d * 81 * , 023514 ( 2010 ) ; g. esposito - farese , c. pitrou and j. p. uzan , phys .
d * 81 * , 063519 ( 2010 ) .
m. a. watanabe , s. kanno and j. soda , phys .
* 102 * , 191302 ( 2009 ) .
s. kanno , j. soda and m. a. watanabe , jcap * 1012 * , 024 ( 2010 ) .
a. e. gumrukcuoglu , b. himmetoglu and m. peloso , phys .
d * 81 * , 063528 ( 2010 ) ; t. r. dulaney and m. i. gresham , phys .
d * 81 * , 103532 ( 2010 ) ; m. a. watanabe , s. kanno and j. soda , prog .
phys . * 123 * , 1041 ( 2010 ) ; j. ohashi , j. soda and s. tsujikawa , jcap * 1312 * , 009 ( 2013 ) ; s. kanno , j. soda and m. a. watanabe , jcap * 0912 * , 009 ( 2009 ) ; m. a. watanabe , s. kanno and j. soda , mon . not .
soc . * 412 * , l83 ( 2011 ) ; s. hervik , d. f. mota and m. thorsrud , jhep * 1111 * , 146 ( 2011 ) ; m. thorsrud , d. f. mota and s. hervik , jhep * 1210 * , 066 ( 2012 ) ; r. emami and h. firouzjahi , jcap * 1310 * , 041 ( 2013 ) ; x. chen , r. emami , h. firouzjahi and y. wang , jcap * 1408 * , 027 ( 2014 ) . p. v. moniz and j. ward , class .
* 27 * , 235009 ( 2010 ) ; k. murata and j. soda , jcap * 1106 * , 037 ( 2011 ) ; r. emami , h. firouzjahi , s. m. sadegh movahed and m. zarei , jcap * 1102 * , 005 ( 2011 ) ; k. yamamoto , m. a. watanabe and j. soda , class . quant .
* 29 * , 145008 ( 2012 ) ; n. bartolo , s. matarrese , m. peloso and a. ricciardone , jcap * 1308 * , 022 ( 2013 ) ; t. q. do and w. f. kao , phys .
d * 84 * , 123009 ( 2011 ) ; t. q. do , w. f. kao and i. -c .
lin , phys .
d * 83 * , 123002 ( 2011 ) ; s. bhowmick and s. mukherji , mod .
a * 27 * , 1250009 ( 2012 ) ; j. ohashi , j. soda and s. tsujikawa , phys .
d * 87 * , no .
8 , 083520 ( 2013 ) ; j. ohashi , j. soda and s. tsujikawa , phys .
d * 88 * , 103517 ( 2013 ) ; r. emami , h. firouzjahi and m. zarei , phys .
d * 90 * , 023504 ( 2014 ) .
a. maleknejad and m. m. sheikh - jabbari , phys .
d * 85 * , 123508 ( 2012 ) . | it is widely believed that anisotropy in the expansion of the universe will decay exponentially fast during inflation .
this is often referred to as the cosmic no - hair conjecture . however , we find a counter example to the cosmic no - hair conjecture in the context of supergravity . as a demonstration , we present an exact anisotropic power - law inflationary solution which is an attractor in the phase space .
we emphasize that anisotropic inflation is quite generic in the presence of anisotropic sources which couple with an inflaton .
kobe - th-14 - 10 |
the @xmath0 model in flat space is a scalar field theory whose configuration space @xmath2 consists of finite energy maps from euclidean @xmath3to the complex projective space @xmath0 , the energy functional being constructed naturally from the riemannian structures of the base and target spaces ( that is , the model is a pure sigma model in the broad sense ) .
the requirement of finite energy imposes a boundary condition at spatial infinity , that the field approaches the same constant value , independent of direction in @xmath3 , so that the field may be regarded as a map from the one point compactification @xmath4 to @xmath0 .
since @xmath5 also , finite energy configurations are effectively maps @xmath6 , the homotopy theory of which is well understood , and the configuration space is seen to consist of disconnnected sectors @xmath7 labelled by an integer @xmath8 , the `` topological charge '' ( degree ) , @xmath9 each configuration is trapped within its own sector because time evolution is continuous .
the lorentz invariant , time - dependent model is not integrable but complete solution of the static problem has been achieved by means of a bogomolnyi argument and the general charge @xmath8 moduli space , the space of charge-@xmath8 static solutions @xmath10 , is known ( that _ all _ static , finite energy solutions of the @xmath0 model saturate the bogomolnyi bound is a non - trivial result @xcite ) .
each static solution within the charge-@xmath8 sector has the same energy ( minimum within that sector and proportional to @xmath8 ) , and @xmath11 is parametrized by @xmath12 parameters ( the moduli ) , so such a moduli space may be thought of as the @xmath13-dimensional level bottom of a potential valley defined on the infinite dimensional charge-@xmath8 sector , @xmath7 . low energy _
dynamics _ may be approximated by motion restricted to this valley bottom , a manifold embedded in the full configuration space , and thus inheriting from it a non - trivial metric induced by the kinetic energy functional .
the approximate dynamic problem is reduced to the geodesic problem with this metric , and has been investigated by several authors @xcite . in the unit - charge sector
one here encounters a difficulty : certain components of the metric are singular and the approximation is ill defined . for example ,
unit - charge static solutions are localized lumps of energy with arbitrary spatial scale , so one of the six moduli of @xmath14 is a scale parameter .
motion which changes this parameter is impeded by infinite inertia in the geodesic approximation , a result in conflict with numerical evidence which suggests that lumps collapse under scaling perturbation @xcite .
this problem should not be present in the model defined on a compact two dimensional physical space .
the obvious choice is the @xmath15-sphere because the homotopic partition of the configuration space carries through unchanged .
also , @xmath16 with the standard metric is conformally equivalent to euclidean @xmath17 , and the static @xmath0 model energy functional is conformally invariant , so the whole flat space static analysis is still valid and all the moduli spaces are known .
however , the kinetic energy functional _ does _ change and induces a new , well defined metric on the unit - charge moduli space . by means of the isometry group derived from the spatial and internal symmetries of the full field theory we can place restrictions on the possible structure of this metric , greatly simplifying its evaluation .
the geodesic problem is still too complicated to be solved analytically in general , but by identifying totally geodesic submanifolds , it is possible to obtain the qualitative features of a number of interesting solutions . in particular ,
the possibilities for lumps travelling around the sphere are found to be unexpectedly varied .
the @xmath0 model on the @xmath15-sphere is defined by the lagrangian @xmath19=\int_{s^{2}}\ , ds\ , \frac{\partial_{\mu}w\partial_{\nu}\bar{w}}{(1+|w|^{2})^{2}}\ , \eta^{\mu\nu}\ ] ] where @xmath20 is a complex valued field , @xmath21 is the invariant @xmath16 measure and @xmath22 are the components of the inverse of the lorentzian metric @xmath23 on r(time)@xmath24(space ) , @xmath25 being the natural metric on @xmath16 .
although the language of the @xmath0 model is analytically convenient , the homotopic classification and physical meaning of the field configurations are more easily visualized if we exploit the well known equivalence to the @xmath26 sigma model @xcite . in the latter ,
the scalar field is a three dimensional isovector constrained to have unit length with respect to the euclidean @xmath27 norm ( @xmath28 ) , that is , the target space is the 2-sphere of unit radius with its natural metric , which we will denote @xmath29 for clarity .
( the suffix refers to `` isospace '' in analogy with the internal space of nuclear physics models . )
the @xmath0 field @xmath20 is then thought of as the stereographic image of in the equatorial plane , projected from the north pole , @xmath30 .
explicitly , @xmath31 and @xmath32 then @xmath19\equiv l_{\sigma}[{\mbox{\boldmath $ \phi$}}]=\frac{1}{4}\int_{s^{2}}\ , ds\ , \partial_{\mu}{\mbox{\boldmath $ \phi$}}\cdot\partial_{\nu}{\mbox{\boldmath $ \phi$}}\ , \eta^{\mu\nu}\ ] ] the familiar @xmath26 sigma model lagrangian . a @xmath20 configuration , then , may be visualized as a distribution of unit length arrows over the surface of the physical 2-sphere @xmath33 .
each smooth map @xmath33@xmath34@xmath29 falls into one of a discrete infinity of disjoint homotopy classes , each class associated with a unique integer which may be thought of as the topological degree of the map ( see , for example @xcite ) , so homotopic partition of the configuration space is built in to the model from the start .
we also choose stereographic coordinates @xmath35 on @xmath33 , in terms of which , @xmath36 where @xmath37 , @xmath35 takes all values in @xmath3 and @xmath38 , @xmath39 , @xmath40 .
the radius of @xmath33 has been normalized to unity .
the invariant measure is , @xmath41 and so , @xmath19 = \int\ , dx\ , dy\ , \frac{1}{(1+|w|^{2})^{2}}\left ( \frac{|\dot{w}|^{2}}{(1+r^{2})^{2}}-\left|\frac{\partial w}{\partial x}\right|^{2 } -\left|\frac{\partial w}{\partial y}\right|^{2 } \right).\ ] ] we identify kinetic energy , @xmath42 = \int\ , \frac{dx\ , dy}{(1+r^{2})^{2}}\frac{|\dot{w}|^{2}}{(1+|w|^{2})^{2}}\ ] ] and potential energy @xmath43 = \int\ , dx\ , dy\ , \frac{1}{(1+|w|^{2})^{2}}\left ( \left|\frac{\partial w}{\partial x}\right|^{2}+ \left|\frac{\partial w}{\partial y}\right|^{2 } \right).\ ] ] note that the potential energy is identical to that for flat space by virtue of the conformal invariance of the static model ( stereographic projection is a conformal transformation ) .
thus the familiar bogomolnyi argument @xcite follows immediately and ( @xmath44 run over @xmath45 and @xmath46 represents the @xmath27 vector product in space ) : @xmath47 , \nonumber \\
\rightarrow v[w ] & = & \frac{1}{4}\int\ , dx\ , dy\ , \partial_{i}{\mbox{\boldmath $ \phi$}}\cdot\partial_{i}{\mbox{\boldmath $ \phi$}}\nonumber \\
& \geq & \frac{1}{2}\left|\int\ , dx\ , dy\ , \left(\frac{\partial{\mbox{\boldmath $ \phi$}}}{\partial x}\times\frac{\partial{\mbox{\boldmath $ \phi$}}}{\partial y}\right)\cdot{\mbox{\boldmath $ \phi$}}\right| = 2\pi|n|,\end{aligned}\ ] ] where @xmath20 is in the degree @xmath8 homotopy class , equality holding if and only if @xmath48 which , on substitution of ( [ eq : phidef ] ) becomes the cauchy - riemann condition for @xmath20 to be an analytic function of @xmath49 ( upper sign ) or @xmath50 ( lower sign ) . the former ( latter )
case corresponds to static solutions of positive ( negative ) degree , and if @xmath20 is single valued with finite degree @xmath8 , then it must be a rational map of degree @xmath8 in @xmath51 if @xmath52 or in @xmath53 if @xmath54 .
we shall deal with the unit charge moduli space , consisting of all rational maps of degree 1 in @xmath51 .
since the configuration space and moduli spaces of the flat space and spherical space models are diffeomorphic , we shall use the same notation ( @xmath55 etc . ) in both cases .
the simplest static unit - charge solution is @xmath56 which we shall call the symmetric hedgehog because its field points radially outwards at all points on @xmath33 .
its energy density is uniformly distributed , so it is not really a lump .
since the static model is conformally invariant , any configuration obtained from this by a mbius transformation must be another point on the moduli space .
in fact the orbit of @xmath57 under the mbius group _ is _ the space of degree 1 rational maps , each map being generated by one and only one group element .
thus we may identify the moduli space with the parameter space of the mbius group .
there is a well known matrix representation of mbius transformations @xcite which we denote thus : @xmath58 where @xmath59 so that @xmath60 . the last condition ensures the invertibilty of the transformation and fixes the degree of @xmath20 at 1 .
the mbius group product becomes matrix multiplication , @xmath61 where the left hand side means @xmath62 in obvious notation .
all matrices differing by a constant factor yield the same configuration , and @xmath60 so when we divide by this scaling equivalence we can choose a unimodular matrix as the representative for each equivalence class .
there are two such matrices possible for each distinct configuration because if @xmath63 is unimodular , so is @xmath64 .
thus @xmath65 is a double cover of the moduli space , which we recover by dividing out the equivalence @xmath66 : the moduli space is @xmath67 .
coincidentally , @xmath65 is also a double cover of the proper orthochronous lorentz group .
the statement that any lorentz transformation may be formed by a unique composition of a boost then a rotation ( or _ vice versa _ ) translates to the existence , for all @xmath68 , of @xmath69 and @xmath70 , a positive definite , unimodular , hermitian @xmath71 matrix ( call this set @xmath72 ) , satisfying @xmath73 both @xmath74 and @xmath70 being unique @xcite .
it follows that the space @xmath65 is locally a product of @xmath75 ( the group manifold of @xmath76 ) and @xmath27 ( the parameter space of @xmath72 ) , a result which generalizes globally , @xmath77 .
we may choose local coordinates on @xmath65 by defining the standard euler angles @xmath78 on @xmath75 , @xmath79 and expanding @xmath70 in terms of pauli matrices , @xmath80 @xmath81 being chosen to ensure the unimodular and postive definite properties : @xmath82 the 3-vector ( modulus @xmath83 ) takes all values in @xmath27 , while @xmath84 $ ] , @xmath85 $ ] and @xmath86 $ ] .
these ranges allow @xmath63 to take all values in the double cover @xmath65 . in analyzing the structure of the metric
, it is convenient to work with @xmath65 , checking that the metric is single valued under the identification of @xmath63 with @xmath64 .
the true moduli space @xmath67 is charted by the same coordinates but with @xmath87 lying in the reduced range @xmath88 $ ] , for @xmath74 is then restricted to the `` upper half '' of @xmath75 .
the chart has a coordinate singularity at @xmath89 and at @xmath90 .
the explicit connexion between a point in @xmath14 and the corresponding static solution will be made in section 5 , below .
field dynamics of the @xmath0 model may be visualized as the dynamics of a point particle with `` position '' @xmath91 moving in an infinite - dimensional configuration space .
a solution @xmath92 of the field equations is thought of as a trajectory in this space , motion on which is determined by metric @xmath93 $ ] and potential @xmath94 $ ] . in the unit - charge sector
, the bogomolnyi argument shows that there is a six - dimensional subspace on which the potential achieves its topological minimum value of @xmath95 , and that any perturbation departing from this subspace must involve increasing @xmath96 .
if a configuration sitting at the bottom of this potential valley is given a small velocity tangential to it then we expect the ensuing time - evolved field to stay close to the valley bottom , for departure from it entails climbing up the valley walls . in the geodesic approximation
@xcite we restrict motion to the valley bottom , assuming that orthogonal modes are insignificant .
thus , at all times @xmath92 is a solution of the _ static model _ , but we allow the moduli @xmath97 , denoted collectively by @xmath98 , to vary with time in accordance with the inherited action principle .
so , @xmath99 and the lagrangian is @xmath100 defining the induced metric , @xmath101 and ignoring the irrelevant constant , the lagrangian is recast as that of a free particle moving on a riemannian manifold with metric @xmath102 : @xmath103 the equations of motion are the geodesic equations . in principle
all we need do is evaluate the integrals of ( [ eq : metricdef ] ) , but these are 21 functions of 6 variables so as it stands this is intractable in practice .
it is profitable to take a more circumspect approach , using symmetries of the model to place restrictions on the structure of @xmath102 .
consider the rotation group @xmath104 acting on @xmath33 and @xmath29 .
the former is the group of spatial rotations ( under which @xmath20 , or equivalently , transforms as a scalar ) while the latter is the group of global internal rotations ( henceforth called `` isorotations '' ) of the field of which @xmath20 is the stereographic image .
any such transformation @xmath105 leaves invariant ( a ) the topological charge , so @xmath105 is a bijection @xmath106 , ( b ) the potential energy , so within @xmath7 static solutions are mapped to other static solutions , @xmath107 , and ( c ) the kinetic energy , which induces the metric on @xmath11 . hence @xmath105 is an isometry of @xmath108 . the @xmath76 subgroup of the mbius group s double cover , @xmath65 , acting via the operation @xmath109 defined by equation ( [ eq : dotdef ] ) is the double cover of the group of rotations of the 2-sphere @xcite considered as operations on the projective plane ( spatial or internal , ie acting on @xmath49 or @xmath20 ) .
thus we find that ( @xmath110 ) , @xmath111 produces an isorotation of the configuration @xmath112 , while @xmath113 produces a spatial rotation , both isometries of the induced metric .
the action of the isorotation on the moduli space is simple : @xmath114 the isometry takes the @xmath76 left multiplication action on @xmath115 while leaving the @xmath27 moduli unchanged . using a technique standard in the analysis of isometries in general relativity @xcite , we change from the
coordinate basis on @xmath116 , to a non - coordinate basis , in this case the left - invariant 1-forms of the lie group @xmath76 .
these may be found by expanding the left - invariant 1-form @xmath117 in terms of a convenient basis of the lie algebra @xmath118 , for example @xmath119 .
explicitly , @xmath120 where @xmath121 if we evaluate the metric at one particular point on @xmath115 , for all possible @xmath122 , we can obtain the metric at all other points on @xmath115 because @xmath115 is the isorotation orbit of our base point , and isorotation is an isometry , so the metric must remain constant ( for each ) over the entire orbit .
`` constant '' means unchanging when considered as a geometric object , not that the components with respect to the original coordinate basis are constant , because the basis vectors themselves transform non - trivially .
the basis of ( [ eq : sigma ] ) _ is _ invariant however , so the metric must be of the form @xmath123 where @xmath124 and each of the component functions is independent of @xmath78 .
let us now consider the spatial rotations : @xmath125 the latter in terms of coordinates is @xmath126 the action of conjugation of the hermitian , traceless matrix @xmath127 by a unitary matrix @xmath128 is well known @xcite it is equivalent to a @xmath104 rotation of : @xmath129 where @xmath130 with components @xmath131 .
the action on the left - invariant 1-forms is similar . under @xmath132 , @xmath133 where @xmath134 is the same @xmath104 matrix defined above . thus both and transform as 3-vectors under spatial rotations and as scalars under isorotations .
the metric must be invariant under spatial rotations also , so the task is to construct from , @xmath135 and the most general possible @xmath136 tensor which is scalar under these rotations .
this is @xmath137 @xmath138@xmath139 being 7 unknown functions of @xmath140 only .
the metric may be restricted still further on consideration of a discrete isometry .
the kinetic energy is invariant under the discrete `` parity '' transformations @xmath141 and @xmath142 .
however , neither is an isometry of the moduli space because each reverses the sign of the topological charge , mapping lumps to anti - lumps .
the composite transformation @xmath143 _ is _ an isometry . using the configuration of ( [ eq : dotdef ] ) , @xmath144 in terms of the moduli , @xmath145 is the transformation , @xmath146 this isometry removes two of the terms in ( [ eq : metric ] ) because under it , @xmath147 and @xmath148 so that @xmath149 .
the remaining five functions of @xmath150 are evaluated by choosing convenient orientations for , positions on @xmath115 and tangent vectors ( velocities ) , then calculating the kinetic energy and comparing with ( [ eq : metric ] ) . repeating this four times it is possible to extract the following ( see figure 1 ) : @xmath151 \\ \nonumber c & = & \frac{\pi}{2}-2\pi s_{1}(\chi ) \\
\nonumber d & = & \frac{\pi}{\lambda^{2}}\left[6s_{1}(\chi)-\frac{1}{2}\right ] \\
\nonumber g & \equiv & a\end{aligned}\ ] ] where , @xmath152 \\
\nonumber s_{2}(\chi ) & = & \frac{\chi}{(\chi^{2}-1)^{3}}\left[\chi^{4}-2\chi^{2}\log\chi^{2}-1\right ] .
\end{aligned}\ ] ] note that @xmath153 is a strictly increasing function of @xmath150 , and that @xmath154 .
there appear to be divergences of the functions @xmath138@xmath155 at @xmath156 , but these are in fact removable singularities , so all the limits of vanishing @xmath150 exist .
although @xmath157 and @xmath155 are negative it is straightforward to show that this metric is positive definite , as of course it must be .
the veracity of the statement @xmath158 is established by explicit calculation , there being no obvious symmetry argument in its favour . in summary
then , the metric is @xmath159
before discussing geodesics of the metric ( [ eq : g ] ) we must describe the connexion between a point on the moduli space @xmath160 and its corresponding field configuration .
consider first the 3-dimensional submanifold defined by @xmath161 , parametrized by @xmath162 .
any point in this subspace may be written as @xmath163 where @xmath164 , @xmath165 and @xmath166 .
the lump represented by @xmath167 is @xmath168 this is a distorted hedgehog with the arrows pulled towards the north pole .
the larger @xmath150 is , so the larger @xmath153 is and the greater is the distortion .
although it is usual to define the position of a lump as the position of maximum energy density , we shall refer to this as a lump of sharpness @xmath150 located at @xmath169 , the antipodal point to the energy density peak which occurs where the arrows are stretched apart .
obviously the motion of any point is trivially mirrored by its antipodal image , so this terminology makes sense .
the lump represented by is @xmath170\odot z \\ \nonumber & = & [ r^{\dagger}({\mbox{\boldmath $ \lambda$}}'\cdot{\mbox{\boldmath $ \tau$}})]\odot(r\odot z)=r^{\dagger}\odot[w'(r\odot z)].\end{aligned}\ ] ] this configuration is formed by first performing a spatial rotation taking the arrow at the old point @xmath51 and placing it at the new point @xmath171 without changing its orientation then performing the inverse isorotation .
the result looks like the arrows have been fixed to @xmath33 which has then been rotated by @xmath134 which , as defined , has the action on @xmath16 equivalent to @xmath172 , _ not
_ @xmath128 , acting on @xmath0 via @xmath109 . that is , if we define @xmath173 to be stereographic projection , @xmath174@xmath0 so that @xmath175 , then @xmath176
. so the lump at the north pole is shifted to @xmath177 .
all other points on the moduli space are on the isorotation orbit of this submanifold , and isorotation , while changing the internal orientation of the lump , does not move the lump around on physical space .
thus we can always interpret @xmath178 as the lump s position , and @xmath150 as parametrizing its sharpness .
the symmetric hedgehog has @xmath156 , and large @xmath150 lumps have taller , narrower energy density peaks than small @xmath150 lumps .
one way of attacking the geodesic problem is to reduce its dimension by identifying totally geodesic submanifolds , that is , choosing initial value problems whose solution is simplified by some symmetry . the easiest method for identifying such submanifolds
is to find fixed point sets of discrete groups of isometries .
any isometry maps geodesics to geodesics , so if there were a geodesic starting off in the fixed point set of the isometry and subsequently deviating from it , this would be mapped under the isometry to another geodesic , identical to the first throughout its length in the fixed point set , but deviating from the set in a different direction .
this violates the uniqueness of solutions of ordinary differential equations , so no such geodesic may exist . if the initial data are a point on the fixed point set and a velocity tangential to it , then the geodesic must remain on the fixed point set for all subsequent time . examining ( [ eq : g ] ) we see that @xmath179 is an isometry .
its fixed point set is @xmath115 , the isorotation orbit of the symmetric hedgehog , on which the metric is @xmath180 the kinetic energy is the rotational energy of a totally symmetric rigid body , moment of inertia @xmath181 .
the solutions are just isorotations of the symmetric hedgehog at constant frequency about some fixed axis . in this case
isorotation is equivalent to spatial rotation because @xmath182 .
a less trivial geodesic submanifold is the fixed point set of the parity transformation described above , @xmath145 .
this is a 3-dimensional manifold , the product of the plane @xmath183 in @xmath27 with the circle @xmath184 , \beta=\gamma=0\}\cup\{\alpha\in[0,\pi ] , \beta=\gamma=\pi\}$ ] in @xmath115 .
the circle is more conveniently parametrized if we temporarily allow @xmath185 the domain @xmath88 $ ] , for it is then @xmath186 , \beta=\gamma=0\}$ ] .
this space contains lumps of arbitrary sharpness located on a great circle through the poles of @xmath33 , each lump having an internal phase , so certain of its geodesics may be candidates for `` travelling lumps . ''
introducing spherical polar coordinates for , @xmath187 the plane @xmath183 is parametrized by @xmath188 where @xmath189 $ ] , again gluing two semicircles together and extending the domain of @xmath190 to cover the whole circle in one go .
the metric on this geodesic submanifold is @xmath191 so the kinetic energy is @xmath192\ ] ] where we have used the cyclicity of @xmath190 and @xmath185 to eliminate @xmath193 and @xmath194 in favour of their constant , canonically conjugate momenta , @xmath195 note that constant @xmath196 ( @xmath197 ) does _ not _ imply constant @xmath193 ( @xmath194 ) , nor does @xmath198 ( @xmath199 ) imply @xmath200 ( @xmath201 ) .
this system can be visualized as a point particle of position dependent mass @xmath202 moving in a potential .
it is the form of the potential which determines the broad qualitative features of its behaviour : @xmath203 as can be seen from figure 2 while @xmath204 is monotonically decreasing , @xmath205 is monotonically increasing .
this allows the possibility of potential minima where the forces @xmath206 outwards ( in the sense of increasing @xmath150 ) and @xmath207 inwards are in stable equilibrium .
it certainly is _ not _ possible if @xmath208 , for then @xmath209 as a whole is monotonically decreasing .
this region in the @xmath210 plane is shown shaded in figure 3 .
whatever the initial conditions on @xmath150 , the lump always moves towards infinite @xmath150 without passing through @xmath156 ( which would correspond to the lump swapping hemispheres ) , reaching the singularity @xmath211 , an infinitely tall , sharp spike , in finite time .
thus @xmath108 is geodesically incomplete .
this result follows from the rapid vanishing of the inertia to sharpening , @xmath212 , in the large @xmath150 limit ( see figure 4 ) .
for example , consider the simple case @xmath213 and let @xmath214 and @xmath215 be strictly positive .
it is easily seen that @xmath216 , the time taken to reach the singular spike is proportional to the following integral : @xmath217 the integrand is finite over the integration range ( even if @xmath218 ) , so if @xmath216 diverges it can only be due to the large @xmath150 behaviour .
but the integrand vanishes like @xmath219 at large @xmath150 , fast enough to ensure convergence .
the inclusion of repulsive potentials can only make matters worse , so this singular behaviour extends to the rest of the shaded area . in the unshaded region
, one can define the positive constant @xmath220 such that @xmath221 then the forms of the functions @xmath222 and @xmath223 ( see figure 5 ) suggest that for each @xmath150 , there is one ( and only one ) value of @xmath224 ( call it @xmath225 ) for which @xmath226 has a minimum at @xmath150 . the equilibrium condition is @xmath227 , so @xmath228 inverting the definition of @xmath224 we find that there are two distinct values of @xmath229 for each @xmath225 .
if @xmath229 takes one of these and @xmath230 then @xmath150 will not subsequently change and hence @xmath194 and @xmath193 will also remain constant , allowing the lump to travel around a great circle on @xmath33with constant speed and shape while undergoing constant frequency isorotation .
the two values are @xmath231 substituting ( [ eq : pdefs ] ) we can find the corresponding pair of stable ratios @xmath232 as functions of @xmath150 , @xmath233 ( see figure 6 ) .
thus , for any lump sharpness @xmath150 and travel speed @xmath193 there are two possible isorotation frequencies @xmath194 which allow stable , uniform travel and these two stability `` branches '' never coincide .
it is interesting to note that @xmath234 meaning that very tall , sharp lumps can travel uniformly with @xmath235 .
motion with constant @xmath150 and @xmath236 is simply constant speed spatial rotation carrying the lump around a great circle .
so when the extent of the lump s structure is negligible relative to the radius of curvature of @xmath33 , it can travel in analogous fashion to a flat - space @xmath0 lump @xcite . since @xmath237 takes all positive values ,
whatever value @xmath224 takes there is an equilibrium @xmath150 .
if @xmath214 is near this value , then ( assuming @xmath238 is not too large ) the shape of the lump will oscillate periodically about the preferred sharpness , and its speed of travel round the sphere will vary with the same period . if @xmath238 _ is _ too large , or the lump is initially much too spread out for its @xmath224 , then it will escape to the singular spike in finite time .
let us examine the concrete example @xmath239 .
figure 7 shows the potential @xmath226 with its minimum and the lump travel speed @xmath193 as functions of @xmath150 .
we imagine a particle of position dependent mass moving in this potential and for simplicity take @xmath230 .
clearly , if we release the particle with @xmath240 , it will move off to infinity , the last case mentioned above .
but if @xmath241 , oscillatory motion ensues .
even here there are two qualitatively different cases , because @xmath242 has an absolute minimum at @xmath243 , a turning point which is only reached if @xmath244 ( or @xmath245 ) . if @xmath246 then the speed of travel oscillates in simple phase with the lump sharpness , going from fast , spread - out lump to slow , sharp lump and back again .
but if @xmath247 or @xmath245 the speed undergoes an extra wobble during the middle of the sharpness cycle , speeding up then slowing down again as it passes through its maximum sharpness .
this case corresponds to lumps whose shape oscillates more acutely .
other interesting geodesic submanifolds are generated by computing the fixed point sets @xmath248 of the isometries @xmath249 , simultaneous isorotation and spatial rotation by @xmath250 about the @xmath251 and @xmath252 axes respectively : @xmath253 thus if @xmath254 must point along the @xmath252-axis .
on @xmath255 , @xmath256 where @xmath257 $ ] , whereas if @xmath258 then @xmath259 where @xmath260 .
it follows that also points along the @xmath252-axis , independent of @xmath251 .
the @xmath261 submanifolds are the images of @xmath255 under @xmath262 isorotations about the three axes , so it suffices to solve the geodesic problem on @xmath255 geodesics on @xmath263 , are then obtained by acting with the appropriate isometry .
the choice of @xmath252 does nt matter , and we choose to study the cylinder @xmath264 consisting of lumps of every sharpness located at the north ( south ) pole if @xmath265 ( @xmath266 ) , arbitrarily rotated about the north - south axis . note that @xmath267=0 $ ] on @xmath268 so `` isorotated '' and `` spatially rotated '' mean the same thing in this case .
the kinetic energy on @xmath268 is @xmath269\ ] ] where once again @xmath270 is the momentum conjugate to @xmath271 , @xmath272 and is constant by virtue of the cyclicity of @xmath271 .
this looks like a particle in one dimension moving in a potential @xmath273 with postion dependent mass . from the potential ( figure
8) we see that all motion is oscillatory and that @xmath274 periodically changes sign . thus a lump set spinning about its own axis will spread out , its rotation slowing , until it is uniformly spread over the sphere , whereupon it will shrink to its mirror image in the opposite hemisphere , regaining its original spin speed as it does so .
the process then reverses and the lump `` bounces '' between antipodal points indefinitely . defining the new coordinate @xmath275 which takes values in a finite open interval @xmath276 symmetric about @xmath277 ,
the metric on @xmath268 becomes @xmath278 where @xmath279 . since @xmath280
, the manifold may be embedded as a surface of revolution in @xmath27 and geodesics on it can be visualized directly .
figure 9 is a sketch of the embedded surface , which is of finite length and sausage - shaped with its ends pinched to infinitely sharp spikes , the tips of which are the points @xmath281 and so are missing .
the coordinates @xmath282 are geodesic orthogonal coordinates : a curve of constant @xmath271 is a geodesic along the length of the cylinder , lying in a plane containing the cylinder s axis , parametrized by arc length @xmath283 , while a curve of constant @xmath283 is a circle of radius @xmath284 , lying in a plane orthogonal to the cylinder s axis .
two such curves always intersect at right angles .
the spinning geodesics described above wind around the cylinder , never reaching the ends ( this would violate conservation of `` angular momentum '' ) but winding back and forth between two circles @xmath285 , @xmath286 which they touch tangentially .
the angle @xmath287 at which the geodesic intersects the circle @xmath277 determines @xmath288 .
when @xmath289 the geodesic stays on the circle , @xmath290 ( a spinning symmetric hedgehog ) and @xmath291 monotonically increases , tending to the supremum @xmath292 as @xmath287 tends to @xmath293 ( @xmath294 is an irrotational geodesic between antipodal singular spikes ) .
note that the geodesic incompleteness already mentioned appears again , this time characterized by the finite length of the cylinder and the missing points @xmath295 ( @xmath281 ) .
the behaviour of isolated topological solitons in flat space is generally rather trivial , whereas , as we have seen , despite the homogeneity of @xmath16 , the motion of a single lump on the sphere is surprisingly complicated .
it does travel on great circles , but while doing so its shape may oscillate in phase with its speed , whose periodic variation is of one of two types depending on the violence of the shape oscillations , or it may collapse to an infinitely tall , thin spike in finite time . a lump sent spinning about its own axis spreads out then re - forms in the opposite hemisphere , endlessly commuting between antipodal points .
the infinities in the unit - charge metric in flat space can be attributed to the lumps polynomial tail - off : the kinetic energy needed to rigidly spin or scale - deform a lump diverges because such motions involve changing the field at spatial infinity .
the @xmath0 model on any compact space should be free of this problem because the kinetic energy , being an integral over a space of finite volume , must be finite provided the kinetic energy density is non - singular .
conversely , one would expect the singularity to persist in the model defined on hyperbolic space .
the flat - space @xmath0 model can be made more `` physical '' by adding a @xmath296-dimensional version of the skyrme term to stabilize against lump collapse , and a potential to stabilize against spread .
the bogomolnyi bound remains valid but unsaturable .
the potential is somewhat arbitrary , but one interesting possibilty @xcite gives a mass to small amplitude travelling waves of the field , termed pions in analogy with the skyrme model , and gives the lump an exponential rather than polynomial tail .
this allows the lump to rotate , a problem if one attempts a collective coordinate approximation to low - energy dynamics along the lines recently proposed in @xcite .
the idea is to restrict the field to the `` bogomonyi regime '' moduli space ( in this the space of static @xmath0 solutions ) , introducing a potential and a perturbed metric ( in @xcite but not @xcite ) to account for the new interactions , which are assumed to be weak .
there seems little hope of perturbing the singular flat - space metric such that rotations become possible , but the problem does not arise on the sphere .
the geodesic approximation could be used to investigate the interaction of two lumps moving on @xmath16 .
right angle scattering in head on collisions emerges naturally from the geodesic approximation of many flat - space models as a consequence of the classical indistinguishability of topological solitons .
it would be interesting to see if there is some analogous behaviour on the sphere .
however , evaluating the two - lump metric could be difficult since the action of the isometry group on the charge-2 moduli space is far less accessible than in the present case .
even in flat space @xcite , the scattering problem is sufficiently complicated to require considerable numerical effort .
* acknowledgments : * i would like to thank richard ward , who suggested this work , and bernd schroers for many useful discussions .
i also acknowledge the financial support of the uk particle physics and astronomy research council . | low - energy dynamics in the unit - charge sector of the @xmath0 model on spherical space ( space - time @xmath1 ) is treated in the approximation of geodesic motion on the moduli space of static solutions , a six - dimensional manifold with non - trivial topology and metric .
the structure of the induced metric is restricted by consideration of the isometry group inherited from global symmetries of the full field theory .
evaluation of the metric is then reduced to finding five functions of one coordinate , which may be done explicitly .
some totally geodesic submanifolds are found and the qualitative features of motion on these described .
epsf = -0.4 in = -0.2 in = 6.6 in |
we classify elliptic fibrations on the singular k3 surface @xmath0 associated to the laurent polynomial @xmath2 in order to compute the nron - severi lattice , the picard number , and other basic properties of an algebraic surface , it is useful to identify an elliptic fibration on the surface .
moreover , in view of different applications , one may be interested in finding all the elliptic fibrations of a certain type .
the fibrations of rank @xmath3 and maximal torsion lead more easily to the determination of the @xmath4series of the variety @xcite .
those of positive rank lead to symplectic automorphisms of infinite order of the variety .
lenstra s elliptic curve method ( ecm ) for finding small factors of large numbers originally used elliptic curves on @xmath5 with a torsion - group of order 12 or 16 and rank @xmath6 on @xmath5 @xcite , @xcite .
one way to obtain infinite families of such curves is to use fibrations of modular surfaces , as explained by elkies @xcite .
if the picard number of a k3 surface is large , there may be an infinite number of elliptic fibrations , but there is only a finite number of fibrations up to automorphisms , as proved by sterk @xcite .
oguiso used a geometric method to classify elliptic fibrations in @xcite .
some years later , nishiyama @xcite proposed a lattice - theoretic technique to produce such classifications , recovering oguiso s results and classifying other kummer and k3 surfaces .
since then , results of the same type have been obtained by various authors @xcite , @xcite , @xcite .
recently , the work of @xcite described three possible classifications of elliptic fibrations on a k3 surface , shining a new light on the meaning of what is a class of equivalence of elliptic fibrations .
in particular , they proposed a @xmath7-classification of elliptic fibrations up to automorphisms of the surface and a @xmath8-classification of the frame lattices of the fibrations . for our k3 surface ,
the two classifications coincide .
thus , it is particularly interesting to exhibit here an @xmath8-classification by the kneser - nishiyama method , since in general it is not easy to obtain the @xmath7-classification . this topic will be explained in detail in section [ s : classtype ] .
section [ s : presentations ] is devoted to a toric presentation of the surface @xmath0 , following ideas of @xcite , based on the classification of reflexive polytopes in dimension 3 .
more precisely , the newton polytope of @xmath0 is in the same class as the reflexive polytope of index 1529 .
since , according to @xcite , there is an @xmath9 action on the vertices of polytope 1529 and its polar dual , there is a symplectic action of @xmath9 on @xmath10 this action will be described on specific fibrations .
one of them gives the transcendental lattice @xmath11 we may use these fibrations to relate @xmath0 to a modular elliptic surface analyzed by beauville in @xcite .
we also describe a presentation of @xmath0 found in @xcite , which represents @xmath0 as a k3 surface with a prescribed abelian symplectic automorphism group .
the main results of the paper are obtained by nishiyama s method and are summarized in section [ s : main ] , theorem @xmath12 the classification up to automorphisms of the elliptic fibrations on @xmath0 is given in table [ table : main result ] .
each elliptic fibration is given with the dynkin diagrams characterizing its reducible fibers and the rank and torsion of its mordell - weil group .
more precisely , we obtained @xmath13 elliptic fibrations on @xmath0 , including 17 fibrations of rank @xmath14 and one of rank @xmath15 . due to the high number of different elliptic fibrations ,
we give only a few cases of computing the torsion .
these cases have been selected to give an idea of the various methods involved .
notice the case of fibrations # 22 and # 22b exhibiting two elliptic fibrations with the same singular fibers and torsion but not isomorphic .
corresponding to these different fibrations we give some particularly interesting weierstrass models ; it is possible to make an exhaustive list .
* acknowledgements * we thank the organizers and all those who supported our project for their efficiency , their tenacity and expertise .
the authors of the paper have enjoyed the hospitality of cirm at luminy , which helped to initiate a very fruitful collaboration , gathering from all over the world junior and senior women , bringing their skill , experience and knowledge from geometry and number theory .
our gratitude goes also to the referee for pertinent remarks and helpful comments .
a.g is supported by firb 2012 `` moduli spaces and their applications '' and by prin 20102011 `` geometria delle variet algebriche '' .
c.s is supported by faperj ( grant e26/112.422/2012 ) .
thanks the nsf - awm travel grant program for supporting her visit to cirm .
let @xmath16 be a smooth complex compact projective surface . a surface @xmath16 is a k3 surface if its canonical bundle and its irregularity are trivial , that is , if @xmath17 and @xmath18 .
[ def : elliptic fibration ] a flat surjective map @xmath19 is called an _ elliptic fibration _ if : + 1 ) the generic fiber of @xmath20 is a smooth curve of genus 1 ; + 2 ) there exists at least one section @xmath21 for @xmath20 . in particular , we choose one section of @xmath20 , which we refer to as the zero section .
we always denote by @xmath22 the class of the fiber of an elliptic fibration and by @xmath23 the curve ( and the class of this curve ) which is the image of @xmath24 in @xmath16 .
the group of the sections of an elliptic fibration @xmath20 is called the mordell weil group and is denoted by @xmath25 .
a generic k3 surface does not admit elliptic fibrations , but if the picard number of the k3 is sufficiently large , it is known that the surface must admit at least one elliptic fibration ( see proposition [ prop : e.f . on x with big rho ] ) . on the other hand
, it is known that a k3 surface admits a finite number of elliptic fibrations up to automorphisms ( see proposition [ prop : finite number of e.f . ] ) .
thus , a very natural problem is to classify the elliptic fibrations on a given k3 surface .
this problem has been discussed in several papers , starting in the eighties .
there are essentially two different ways to classify elliptic fibrations on k3 surfaces described in @xcite and @xcite . in some particular cases
, a third method can be applied ; see @xcite .
first , however , we must introduce a different problem : what does it mean to ` classify ' elliptic fibrations ? " a deep and interesting discussion of this problem is given in @xcite , where the authors introduce three different types of classifications of elliptic fibrations and prove that under certain ( strong ) conditions these three different classifications collapse to a unique one .
we observe that it was already known by @xcite that in general these three different classifications do not collapse to a unique one .
we now summarize the results by @xcite and the types of classifications . in this section
we recall some of the main results on elliptic fibrations on k3 surfaces ( for example , compare @xcite ) , and we introduce the different classifications of elliptic fibrations discussed in @xcite .
let @xmath16 be a k3 surface and @xmath19 be an elliptic fibration on @xmath16 .
let @xmath28 be the class of the fiber of @xmath20 .
then @xmath22 is a nef divisor which defines the map @xmath29 which sends every point @xmath30 to @xmath31 , where @xmath32 is a basis of @xmath33 , i.e. a basis of sections of the line bundle associated to the divisor @xmath22 .
the map @xmath34 is the elliptic fibration @xmath20 .
hence , every elliptic fibration on a k3 surface is uniquely associated to an irreducible nef divisor ( with trivial self intersection ) . since @xmath19 admits a section , there exists a rational curve which intersects every fiber in one point .
its class in @xmath35 is denoted by @xmath23 and has the following intersection properties @xmath36 ( since @xmath23 is a rational curve ) and @xmath37 ( since @xmath23 is a section ) . thus , the elliptic fibration @xmath19 ( with a chosen section , as in definition [ def : elliptic fibration ] ) is uniquely associated to a pair of divisors @xmath38 .
this pair of divisors spans a lattice which is isometric to @xmath26 , represented by the matrix @xmath39 $ ] , ( considering the basis @xmath22 , @xmath40 ) .
hence each elliptic fibration is associated to a chosen embedding of @xmath26 in @xmath35 . on the other hand the following result holds : ( ( * ? ? ?
* lemma 2.1 ) and ( * ? ? ?
* corollary 1.13.15))[prop : e.f . on x with big rho ] let @xmath16 be a k3 surface , such that there exists a primitive embedding @xmath41 . then @xmath16 admits an elliptic fibration .
let @xmath16 be a k3 surface with picard number @xmath42 .
then , there is a primitive embedding of @xmath26 in @xmath35 and hence @xmath16 admits at least one elliptic fibration .
a canonical embedding of @xmath43 in @xmath35 is defined as follows : let us denote by @xmath44 and @xmath45 the unique two primitive vectors of @xmath26 with trivial self intersection .
an embedding of @xmath26 in @xmath35 is called _ canonical _ if the image of @xmath44 in @xmath35 is a nef divisor and the image of @xmath46 in @xmath35 is an effective irreducible divisor . the first naive classification of the elliptic fibrations that one can consider is the classification described above , roughly speaking : two fibrations are different if they correspond to different irreducible nef divisors with trivial self intersections .
this essentially coincides with the classification of the canonical embeddings of @xmath26 in @xmath35 . following @xcite we call this classification the @xmath27-classification of the elliptic fibrations on @xmath16 .
clearly , it is possible ( and indeed likely , if the picard number is sufficiently large ) that there is an infinite number of irreducible nef divisors with trivial self intersection and also infinitely many copies of @xmath26 canonically embedded in @xmath35 .
thus , it is possible that there is an infinite number of fibrations in curves of genus 1 on @xmath16 and moreover an infinite number of elliptic fibrations on @xmath16 .
the automorphism group of a variety transforms the variety to itself preserving its structure , but moves points and subvarieties on the variety .
thus , if one is considering a variety with a nontrivial automorphism group , one usually classifies objects on the variety up to automorphisms .
let @xmath16 be a k3 surface with a sufficiently large picard number ( at least 2 ) .
then the automorphism group of @xmath16 is in general nontrivial , and it is often of infinite order .
more precisely , if @xmath48 , then the automorphism group of @xmath16 is finite if and only if there is a vector with self intersection either 0 or @xmath49 in the nron severi group . if @xmath50 , then the automorphism group of @xmath16 is finite if and only if the nron
severi group is isometric to a lattice contained in a known finite list of lattices , cf .
let us assume that @xmath16 admits more than one elliptic fibration ( up to the @xmath27-classification defined above ) .
this means that there exist at least two elliptic fibrations @xmath19 and @xmath51 such that @xmath52 , where @xmath22 ( resp .
@xmath53 ) is the class of the fiber of the fibration @xmath20 ( resp .
@xmath54 ) . by the previous observation
, it seems very natural to consider @xmath20 and @xmath54 equivalent if there exists an automorphism of @xmath16 which sends @xmath20 to @xmath55 .
this is the idea behind the @xmath47-classification of the elliptic fibrations introduced in @xcite .
the @xmath47-classification of the elliptic fibrations on a k3 surface is the classification of elliptic fibrations up to automorphisms of the surface . to be more precise : @xmath20 is @xmath47-equivalent to @xmath54 if and only if there exists @xmath56 such that @xmath57 .
we observe that if two elliptic fibrations on a k3 surface are equivalent up to automorphism , then all their geometric properties ( the type and the number of singular fibers , the properties of the mordell weil group and the intersection properties of the sections ) coincide .
this is true essentially by definition , since an automorphism preserves all the `` geometric '' properties of subvarieties on @xmath16 .
the advantages of the @xmath47-classification with respect to the @xmath27-classification are essentially two .
the first is more philosophical : in several contexts , to classify an object on varieties means to classify the object up to automorphisms of the variety .
the second is more practical and is based on an important result by sterk : the @xmath47-classification must have a finite number of classes : @xcite[prop : finite number of e.f . ] up to automorphisms , there exists a finite number of elliptic fibrations on a k3 surface .
the main problem of the @xmath47-classification is that it is difficult to obtain a @xmath47-classification of elliptic fibrations on k3 surfaces , since it is in general difficult to give a complete description of the automorphism group of a k3 surface and the orbit of divisors under this group .
an intermediate classification can be introduced , the @xmath58-classification .
the @xmath58-classification is not as fine as the @xmath47-classification , and its geometric meaning is not as clear as the meanings of the classifications introduced above .
however , the @xmath58-classification can be described in a very natural way in the context of lattice theory , and there is a standard method to produce it . since the @xmath58-classification is essentially the classification of certain lattices strictly related to the elliptic fibrations , we recall here some definitions and properties of lattices related to an elliptic fibration .
we have already observed that every elliptic fibration on @xmath16 is associated to an embedding @xmath59 .
the orthogonal complement of @xmath60 in @xmath35 , @xmath61 , is denoted by @xmath62 and called the frame lattice of @xmath20 . the frame lattice of @xmath20 encodes essentially all the geometric properties of @xmath20 , as we explain now .
we recall that the irreducible components of the reducible fibers which do not meet the zero section generate a root lattice , which is the direct sum of certain dynkin diagrams .
let us consider the root lattice @xmath63 of @xmath62 .
then the lattice @xmath63 is exactly the direct sum of the dynkin diagram corresponding to the reducible fibers . to be more precise if the lattice @xmath64 ( resp .
@xmath65 , @xmath66 , @xmath67 , @xmath68 , @xmath69 , @xmath70 ) is a summand of the lattice @xmath63 , then the fibration @xmath20 admits a fiber of type @xmath71 ( resp .
@xmath72 , @xmath73 , @xmath74 , @xmath75 ) .
however , the lattices @xmath76 and @xmath77 can be associated to two different types of reducible fibers , i.e. to @xmath78 and @xmath79 and to @xmath80 and @xmath81 respectively .
we can not distinguish between these two different cases using lattice theory .
moreover , the singular non - reducible fibers of an elliptic fibration can be either of type @xmath82 or of type @xmath83 .
given an elliptic fibration @xmath20 on a k3 surface @xmath16 , the lattice @xmath84 is often called the _ trivial lattice _ ( see ( * ? ? ?
* lemma 8.3 ) for a more detailed discussion ) .
let us now consider the mordell weil group of an elliptic fibration @xmath20 on a k3 surface @xmath16 : its properties are also encoded in the frame @xmath62 , indeed @xmath85 . in particular , @xmath86 @xmath87 where , for every sublattice @xmath88 , @xmath89 denotes the primitive closure of @xmath90 in @xmath35 , i.e.@xmath91 .
the @xmath58-classification of elliptic fibrations on a k3 surface is the classification of their frame lattices .
it appears now clear that if two elliptic fibrations are identified by the @xmath58-classification , they have the same trivial lattice and the same mordell
weil group ( since these objects are uniquely determined by the frame of the elliptic fibration ) .
we observe that if @xmath20 and @xmath54 are identified by the @xmath47-classification , then there exists an automorphism @xmath92 , such that @xmath57 .
the automorphism @xmath93 induces an isometry @xmath94 on @xmath35 and it is clear that @xmath95 is an isometry . thus the elliptic fibrations @xmath20 and @xmath54 have isometric frame lattices and so are @xmath58-equivalent . the @xmath58-classification is not as fine as the @xmath47-classification ; indeed , if @xmath96 is an isometry , a priori there is no reason to conclude that there exists an automorphism @xmath92 such that @xmath97 ; indeed comparing the @xmath47-classification given in @xcite and the @xmath58-classification given in @xcite for the kummer surface of the product of two non - isogenous elliptic curves , one can check that the first one is more fine than the second one .
the advantage of the @xmath58-classification sits in its strong relation with the lattice theory ; indeed , there is a method which allows one to obtain the @xmath58-classification of elliptic fibration on several k3 surfaces .
this method is presented in @xcite and will be described in this paper in section [ sec : details ] .
one of the main results of @xcite is about the relations among the various types of classifications of elliptic fibrations on k3 surfaces .
first we observe that there exists two surjective maps @xmath98 and @xmath99 , which are in fact quotient maps ( cf .
* formulae ( 54 ) and ( 57 ) ) ) .
this induces a map @xmath100 which is not necessarily a quotient map .
the ( * ? ? ?
* proposition c ) gives a bound for the number of different elliptic fibrations up to the @xmath47-classification , which are identified by the @xmath58-classification . as a corollary the following is proved : [ cor : j2=j1 ] ( ( * ? ? ?
* corollary d ) ) let @xmath101 be a k3 surface such that the transcendental lattice of @xmath16 is isometric to @xmath102 $ ] .
if @xmath103 is one of the following @xmath104 , @xmath105 , @xmath106 , @xmath107 , @xmath108 , @xmath109 , @xmath110 , @xmath111 , @xmath112 , @xmath113 , then @xmath114 .
the first paper about the classification of elliptic fibrations on k3 surfaces is due to oguiso , @xcite .
he gives a @xmath47-classification of the elliptic fibrations on the kummer surface of the product of two non - isogenous elliptic curves .
the method proposed in @xcite is very geometric : it is strictly related to the presence of a certain automorphism ( a non
symplectic involution ) on the k3 surface . since one has to require that the k3 surface admits this special automorphism , the method suggested in @xcite can be generalized only to certain special k3 surfaces ( see @xcite and @xcite ) .
seven years after the paper @xcite , a different method was proposed by nishiyama in @xcite .
this method is less geometric and more related to the lattice structure of the k3 surfaces and of the elliptic fibrations .
nishiyama applied this method in order to obtain a @xmath58-classification of the elliptic fibrations , both on the k3 surface already considered in @xcite and on other k3 surfaces ( cyclic quotients of the product of two special elliptic curves ) to which the method by oguiso can not be applied . later , in @xcite ,
the method is used to give a @xmath58-classification of elliptic fibrations on a k3 surface whose transcendental lattice is @xmath115 .
the main idea of nishiyama s method is the following : we consider a k3 surface @xmath16 and its transcendental lattice @xmath116 .
then we consider a lattice @xmath117 such that : @xmath117 is negative definite ; @xmath118 ; the discriminant group and form of @xmath117 are the same as the ones of @xmath116 .
we consider primitive embeddings of @xmath119 , where @xmath90 is a niemeier lattice .
the orthogonal complement of @xmath120 in @xmath90 is in fact the frame of an elliptic fibration on @xmath16 .
the classification of the primitive embeddings of @xmath117 in @xmath90 for every niemeier lattice @xmath90 coincides with the @xmath58-classification of the elliptic fibrations on @xmath16
. we will give more details on nishiyama s method in section [ sec : details ] .
since this method is related only to the lattice properties of the surface , a priori one can not expect to find a @xmath47-classification by using only this method .
thanks to corollary [ cor : j2=j1 ] , ( see @xcite ) the results obtained by nishiyama s method are sometimes stronger than expected . in particular , we will see that in our case ( as in the case described in @xcite ) the classification that we obtain for the elliptic fibrations on a certain k3 surface using the nishiyama s method , is in fact a @xmath47-classification ( and not only a @xmath58-classification ) . in section [ sec : explicit computations ] , we will classify elliptic fibrations on a certain k3 surface , determining both the trivial lattice and the mordell weil group .
a priori , steps ( 8) and ( 9 ) of the algorithm presented in [ sec : details ] completely determine the mordell weil group . in any case
, we can deduce some information on the torsion part of the mordell weil group by considering only the properties of the reducible fibers of the elliptic fibration .
this makes the computation easier , so here we collect some results on the relations between the reducible fibers of a fibration and the torsion part of the mordell weil group .
first , we recall that a section meets every fiber in exactly one smooth point , so a section meets every reducible fiber in one point of a component with multiplicity 1 ( we recall that the fibers of type @xmath121 , @xmath71 , @xmath73 , @xmath72 have reducible components with multiplicity greater than 1 )
. we will call the component of a reducible fiber which meets the zero section the _ zero component _ or _
trivial component_. every section ( being a rational point of an elliptic curve defined over @xmath122 ) induces an automorphism of every fiber , in particular of every reducible fiber .
thus , the presence of an @xmath123-torsion section implies that all the reducible fibers of the fibration admits @xmath124 as subgroup of the automorphism group .
in particular , this implies the following ( well known ) result : ( cf .
* section 7.2))[prop : torsion and reducible fibers ] let @xmath19 be an elliptic fibration and let @xmath125 the torsion part of the mordell weil group . if there is a fiber of type @xmath71 , then @xmath126 .
if there is a fiber of type @xmath73 , then @xmath127 . if there is a fiber of type @xmath72 , then @xmath128
if there is a fiber of type @xmath121 and @xmath123 is an even number , then @xmath129 .
if there is a fiber of type @xmath121 and @xmath123 is an odd number , then @xmath130
. the theory of universal elliptic surfaces parametrizing elliptic curves with prescribed torsion can also be useful when finding the torsion subgroup of a few elliptic fibrations on the list .
it relies on the following definition / proposition .
let @xmath131 be an elliptic fibration on a surface @xmath0 .
assume @xmath132 has a section of order @xmath133 , for some @xmath134 , with @xmath135 .
then @xmath0 is a cover of the universal modular elliptic surface , @xmath136 of level @xmath133 . after studying the possible singular fibers of the universal surfaces above
, one gets the following .
let @xmath137 be the universal modular elliptic surface of level @xmath138 .
the following hold : + i ) if @xmath139 then @xmath137 admits only semi - stable singular fibers .
they are all of type @xmath140 with @xmath141 .
+ ii ) the surface @xmath142 is a rational elliptic surface with singular fibers @xmath143 .
+ the group structure of the mordell weil group is the group structure of the rational points of the elliptic curve defined over the function field of the basis of the fibration .
it is also possible to equip the mordell weil group of a pairing taking values in @xmath144 , which transforms the mordell weil group to a @xmath144-lattice .
here we recall the definitions and the main properties of this pairing . for a more detailed description
we refer to @xcite and to the original paper @xcite .
let @xmath145 be an elliptic fibration and let @xmath23 be the zero section .
the height pairing is the @xmath144-valued pairing , @xmath146 defined on the sections of an elliptic fibration as follows : @xmath147 where @xmath148 is the holomorphic characteristic of the surface @xmath16 , @xmath149 is the intersection form on @xmath35 , @xmath150 and @xmath151 is a contribution which depends on the type of the reducible fiber and on the intersection of @xmath152 and @xmath153 with such a fiber as described in ( * ? ? ?
* table 4 ) . the value @xmath154 is called the _ height _ of the section @xmath152 .
we observe that the height formula is induced by the projection of the intersection form on @xmath155 to the orthogonal complement of the trivial lattice @xmath156 ( cf .
* section 11 ) ) .
( ( * ? ? ?
* section 11.6))[prop : height and torsion ] let @xmath157 be a section of the elliptic fibration @xmath145 .
the section @xmath152 is a torsion section if and only if @xmath158 .
the goal of this paper is the classification of the elliptic fibrations on the unique k3 surface @xmath0 such that @xmath159 .
this surface is interesting for several reasons , and we will present it from different points of view .
let @xmath133 be a lattice isomorphic to @xmath161 .
the dual lattice @xmath162 of @xmath133 is given by @xmath163 ; it is also isomorphic to @xmath161 .
we write the pairing of @xmath164 and @xmath165 as @xmath166 . given a lattice polytope @xmath167 in @xmath133 ,
we define its _ polar polytope _ @xmath168 to be @xmath169m@xmath170 .
if @xmath168 is also a lattice polytope , we say that @xmath167 is a reflexive polytope and that @xmath167 and @xmath168 are a mirror pair .
a reflexive polytope must contain @xmath171 ; furthermore , @xmath171 is the only interior lattice point of the polytope .
reflexive polytopes have been classified in 1,2,3 , and 4 dimensions . in 3 dimensions ,
there are 4,319 reflexive polytopes , up to an overall isomorphism preserving lattice structure @xcite .
the database of reflexive polytopes is incorporated in the open - source computer algebra software @xcite .
now , consider the one - parameter family of k3 surfaces given by @xmath172 this family of k3 surfaces was first studied in @xcite , where its picard - fuchs equation was computed . a general member of the family has picard lattice given by @xmath173 .
the newton polytope @xmath168 determined by the family of polynomials in equation [ e : affineverrillpencil ] is a reflexive polytope with 12 vertices and 14 facets .
this polytope has the greatest number of facets of any three - dimensional reflexive polytope ; furthermore , there is a unique three - dimensional reflexive polytope with this property , up to isomorphism . in the database of reflexive polytopes found in @xcite ,
this polytope has index 1529 .
we illustrate @xmath168 and its polar polytope @xmath167 in figures [ f:2355 ] and [ f:1529 ] . 2
let us recall some standard constructions and notations involving toric varieties .
a _ cone _ in @xmath133 is a subset of the real vector space @xmath174 generated by nonnegative @xmath175-linear combinations of a set of vectors @xmath176 .
we assume that cones are strongly convex , that is , they contain no line through the origin .
note that each face of a cone is a cone .
a _ fan _ @xmath177 consists of a finite collection of cones such that each face of a cone in the fan is also in the fan , and any pair of cones in the fan intersects in a common face .
we say @xmath177 is _ simplicial _ if the generators of each cone in @xmath177 are linearly independent over @xmath175 .
if every element of @xmath178 belongs to some cone in @xmath177 , we say @xmath177 is _
complete_. a fan @xmath177 defines a toric variety @xmath179 .
if the fan is complete , we may describe @xmath179 using homogeneous coordinates , in a process analogous to the construction of @xmath180 as a quotient space of @xmath181 .
the homogeneous coordinates have one coordinate @xmath182 for each generator of a one - dimensional cone of @xmath177 . we may obtain a fan @xmath183 from a mirror pair of reflexive polytopes in two equivalent ways
. we may take cones over the faces of @xmath184 , or we may take the _ normal fan _ to the polytope @xmath185 .
let @xmath177 be a simplicial refinement of @xmath183 such that the one - dimensional cones of @xmath177 are generated by the nonzero lattice points @xmath186 , @xmath187 , of @xmath167 ; we call such a refinement a _ maximal projective subdivision_. then the variety @xmath179 is an orbifold . then in homogeneous coordinates , we have one coordinate @xmath188 for each nonzero lattice point in @xmath167 .
we may describe the anticanonical hypersurfaces in homogeneous coordinates using polynomials of the form : @xmath189 here the @xmath190 are arbitrary coefficients .
note that @xmath191 has one monomial for each lattice point of @xmath168 .
if the reflexive polytope @xmath167 is three - dimensional , @xmath179 is smooth and smooth anticanonical hypersurfaces in @xmath179 are k3 surfaces ( see @xcite for details ) .
the orientation - preserving symmetry group of @xmath167 and @xmath168 is the symmetric group @xmath160 .
this group acts transitively on the vertices of @xmath168 .
as the authors of @xcite observe , by setting the coefficients @xmath190 corresponding to the vertices of @xmath168 to 1 and the coefficient corresponding to the origin to a parameter @xmath192 , we obtain a naturally one - parameter family of k3 hypersurfaces with generic picard rank 19 : @xmath193 equation [ e : verrillpencil ] is simply equation [ e : affineverrillpencil ] in homogeneous coordinates .
if we view @xmath160 as acting on the vertices of @xmath167 rather than the vertices of @xmath168 , we obtain a permutation of the homogeneous coordinates @xmath188 .
the authors of @xcite show that this action of @xmath160 restricts to a symplectic action on each k3 surface in the pencil given by equation [ e : verrillpencil ] ; in particular , we have a symplectic action of @xmath160 on @xmath0 . in the affine
coordinates of equation [ e : affineverrillpencil ] , the group action is generated by an element @xmath194 of order 2 which acts by @xmath195 and an element @xmath196 of order 4 which acts by @xmath197 .
let @xmath0 be the k3 surface defined by @xmath198 , where @xmath22 is the numerator of @xmath199 the k3 surface @xmath0 is the special member of the family of k3 surfaces described in which is obtained by setting @xmath200 .
+ we will use three elements of the symplectic group @xmath160 : the three - cycle @xmath201 given by @xmath202 the four - cycle @xmath203 and the two - cycle @xmath204 we describe explicitly a first elliptic fibration , which gives the main properties of @xmath10 we use the following factorizations @xmath205 if @xmath206 we see that @xmath207 is invariant under the action of @xmath208 if we substitute @xmath209 for @xmath210 we obtain the equation of an elliptic curve , so the morphism @xmath211 is an elliptic fibration of @xmath0 .
we use the birational transformation@xmath212 with inverse@xmath213 to obtain the weierstrass equation @xmath214 notice the torsion points @xmath215 and @xmath216 of order @xmath15 and the @xmath15 points of order @xmath14 with @xmath217coordinate @xmath218,@xmath219 , and @xmath220 we use also the weierstrass form@xmath221 with@xmath222 the singular fibers are of type @xmath223 for @xmath224 and @xmath225 for @xmath226 so the trivial lattice of this fibration is @xmath227 . hence the picard number of @xmath0 is @xmath228 and @xmath229 .
so @xmath0 is a singular k3 surface .
this elliptic fibration is contained in the ( * ? ? ? * table 2 line 4 ) and thus its transcendental lattice is @xmath230
. moreover , all the fibers have split mulitiplicative reduction and thus the nron severi group is generated by curves defined on @xmath5 .
we have already observed that @xmath0 is a special member of the 1-dimensional family of k3 surfaces defined by equation [ e : affineverrillpencil ] .
indeed , the transcendental lattice of @xmath0 is primitively embedded in @xmath231 by the vectors @xmath232 .
this gives the following proposition : [ prop : the surface x ] the nron - severi group of the k3 surface @xmath0 has rank 20 and is generated by divisors which are defined over @xmath5 .
the transcendental lattice of @xmath0 is @xmath233{cc}2 & 0\\ 0 & 6 \end{array } \right).$ ] the equation is the universal elliptic curve with torsion group @xmath234 and is in fact equivalent to the equation given in @xcite.thus this fibration can be called modular : we can view the base curve @xmath235 as the modular curve @xmath236 with @xmath237 @xmath238{cc}a & b\\ c & d \end{array } \right ) \in sl_{2}\left ( \mathbb{z}\right ) , a\equiv1\operatorname{mod}6,c\equiv0\operatorname{mod}6,b\equiv0\operatorname{mod}2\}$ ] . by we
can easily obtain the equation @xmath239 and realize @xmath0 by a base change of the modular rational elliptic surface @xmath240 described by beauville in @xcite .
we can prove that on the fiber , the automorphism @xmath201 corresponds to adding a @xmath15-torsion point .
if @xmath241 we see that @xmath242 is invariant under the action of @xmath243 substituting @xmath244 for @xmath245 in @xmath22 , we obtain the equation of an elliptic curve .
using standard transformations ( as in @xcite(39.2 ) , @xcite or @xcite ) we obtain the weierstrass model @xmath246 the point @xmath247 is of order @xmath248 the point @xmath249 is of infinite order .
the singular fibers are @xmath250 one can prove that on the fiber , @xmath203 corresponds to the translation by a @xmath251torsion point .
moreover , the translation by the point @xmath252 defines an automorphism of infinite order on @xmath0 .
if we compute the height of @xmath252 we can show , using shioda formula , @xcite that @xmath252 and @xmath253 generate the mordell - weil group .
if @xmath255 , we see that @xmath256 is invariant under the action of @xmath204 substituting @xmath257 for @xmath245 we obtain the equation of an elliptic curve and the following weierstrass model @xmath258 the point @xmath259 is a two - torsion point .
the point @xmath260 is of infinite order .
the singular fibers are @xmath261 .
[ rem : elliptic fibration elkies]from elkies results @xcite @xcite there is a unique k3 surface @xmath262 with nron - severi group of rank @xmath228 and discriminant @xmath263 that consists entirely of classes of divisors defined over @xmath5 .
indeed it is @xmath0 .
moreover , in @xcite a weierstrass equation for an elliptic fibration on @xmath0 defined over @xmath144 , is given : @xmath264 the surface @xmath0 is considered also in a slightly different context in @xcite because of its relation with the study of the moduli space of k3 surfaces with a symplectic action of a finite abelian group .
indeed , the aim of the paper @xcite is to study elliptic fibrations @xmath265 on k3 surfaces @xmath266 such that @xmath267 is a torsion group . since the translation by a section is a symplectic automorphism of @xmath266 , if @xmath267
, then @xmath268 is a group which acts symplectically on @xmath266 .
in @xcite it is shown how one can describe both a basis for the nron
severi group of @xmath266 and the action induced by the symplectic action of @xmath268 on this basis . in particular , one can directly compute the lattices @xmath269 and @xmath270 .
the latter does not depend on @xmath266 but only on @xmath268 and its computation plays a central role in the description of the moduli space of the k3 surfaces admitting a symplectic action of @xmath268 ( see @xcite and @xcite ) .
in particular , the case @xmath271 is considered : in this case the k3 surface @xmath266 is @xmath0 , and the elliptic fibration @xmath272 is . comparing the symplectic action of @xmath273 on @xmath0 given in @xcite with the symplectic group action of @xmath160 described in [ ss : torichypersurface ] , we find that the two groups intersect in the subgroup of order 3 generated by the map @xmath274 given by @xmath275 .
this section is devoted to the proof of our main result : the classification up to automorphisms of the elliptic fibrations on @xmath0 is given in table [ table : main result ] .
each elliptic fibration is given with the dynkin diagrams characterizing its reducible fibers and the rank and torsion of its mordell - weil group .
more precisely , we obtained @xmath13 elliptic fibrations on @xmath0 , including 17 fibrations of rank @xmath14 and one of rank @xmath15 . in order to compute @xmath278
we recall that @xmath278 is an overlattice of finite index of @xmath133 ; in fact , it contains the non trivial elements of @xmath279 which are orthogonal to @xmath280 .
moreover , the index of the inclusion @xmath281 depends on the discriminant of @xmath133 .
indeed @xmath282 , so the index of the inclusion @xmath281 is @xmath283 . as example
we compute here the lattices @xmath278 for the two different embeddings of @xmath284 in @xmath285 ( i.e. for the fibrations 22 and 22(b ) ) .
thus , we consider the niemeier lattice @xmath90 such that @xmath286 and we denote the generators of @xmath279 as follows : @xmath287 , @xmath288 , @xmath289 * fibration @xmath290 * we consider the embedding @xmath291 such that @xmath292 .
the generators of the lattice @xmath133 are described in table [ table : orthogonal ] and one can directly check that @xmath293 .
so , @xmath294 and the index of the inclusion @xmath281 is @xmath295 .
this implies that there is a copy of @xmath296 which is also contained in @xmath278 and so in particular is orthogonal to @xmath297 .
we observe that @xmath298 is orthogonal to the embedded copy of @xmath284 , @xmath299 and @xmath300 are not .
moreover @xmath301 is orthogonal to the embedded copy of @xmath284 .
hence @xmath298 and @xmath301 generates @xmath302 .
we just observe that @xmath303 is equivalent mod @xmath304 to the vector @xmath305 , so @xmath306 .
we will reconsider this fibration in section [ sec : again on 22 and 22b ] comparing it with the fibration @xmath307 .
+ * fibration @xmath308 * we consider the other embedding of @xmath284 in @xmath309 , i.e. @xmath310 such that @xmath311 the generators of the lattice @xmath133 is described in table [ table : orthogonal ] and one can directly check that @xmath312 . as above this implies that @xmath302 which is generated by elements in @xmath279 which are orthogonal to @xmath313 .
in particular , @xmath314 and @xmath301 are orthogonal to @xmath313 so @xmath315 and @xmath303 .
moreover , @xmath316 .
so , denoted by @xmath317 , we have that @xmath318 .
we will reconsider this fibration in section [ sec : again on 22 and 22b ] comparing it with the fibration @xmath319 .
we recalled in section [ sec : basic on elliptic fibrations ] that each elliptic fibration is associated to a certain decomposition of the nron
severi group as a direct sum of @xmath26 and a lattice , called @xmath278 . in step 5
we computed all the admissible lattices @xmath278 , so we classify the elliptic fibrations on @xmath0 .
we denote all the elliptic fibrations according to their associated embeddings ; this gives the first five columns of the table [ table : main result ] .
moreover , again in section [ sec : basic on elliptic fibrations ] , we recalled that each reducible fiber of an elliptic fibration is uniquely associated to a dynkin diagram and that a dynkin diagram is associated to at most two reducible fibers of the fibration .
this completes step 7 . in order to compute the torsion part of the mordell weil group one has to identify the vectors @xmath323 such that @xmath324 for a certain nontrivial integer number @xmath325 ; this gives the last column of table [ table : main result ] .
we will demonstrate this procedure in some examples below ( on fibrations @xmath319 and @xmath326 ) , but first we remark that in several cases it is possible to use an alternative method either in order to completely determine @xmath125 or at least to bound it .
we already presented the theoretical aspect of these techniques in section [ subset : torsion mw theory ] .
probably the easiest case is the one where @xmath327 . in this case @xmath328 . since @xmath327
, this implies that @xmath329 , so @xmath330 .
hence @xmath331 , thus every element @xmath332 is such that a multiple is contained in @xmath321 , i.e. every element of @xmath333 contributes to the torsion .
thus , @xmath334 .
this immediately allows to compute the torsion for the 7 extremal fibrations @xmath335 * fibration @xmath342 ( fibers of special type , proposition [ prop : torsion and reducible fibers ] ) . *
the presence of the lattice @xmath64 as summand of @xmath321 implies that the fibration has a fiber of type @xmath71 ( two in this specific case ) .
hence @xmath125 is trivial .
* fibration @xmath343 ( fibers of special type , proposition [ prop : torsion and reducible fibers ] ) .
* by proposition [ prop : torsion and reducible fibers ] if a fibration has a fiber of type @xmath72 , then the mordell weil group is a subgroup of @xmath344 . on the other hand , a fiber of type @xmath345 , i.e.
, @xmath346 can only occur in fibrations with 4 or 2-torsion or trivial torsion group .
therefore @xmath125 is trivial .
* fibration @xmath347 ( the height formula , section [ subset : height ] ) .
* suppose there is a non - trivial torsion section @xmath152 .
then , taking into account the possible contributions of the reducible fibers to the height pairing , there is @xmath348 such that one of the following holds : @xmath349 after a simple calculation , one sees that neither of the above can happen and therefore the torsion group @xmath125 is trivial
. * fibration @xmath350 .
* we already computed the generators of @xmath333 in section [ sub sub : step 5 ] , @xmath306 .
a basis of @xmath321 is @xmath351 .
so @xmath352 and @xmath353 since @xmath354 is not a multiple of @xmath355 .
viceversa @xmath356 thus @xmath357 .
as we can check in table [ table : main result ] and we proved in the previous sections , the fibrations @xmath319 and @xmath362 are associated to the same lattice @xmath133 and to the same mordell
weil group .
however , we proved in lemma [ lemma : not unique in d8 ] that they are associated to different ( up to weyl group ) embeddings in the niemeier lattices , so they correspond to fibrations which are not identified by the @xmath58-fibration and in particular they can not have the same frame .
the following question is now natural : what is the difference between these two fibrations ?
the answer is that the section of infinite order , which generates the free part of the mordell weil group of these two fibrations , has different intersection properties , as we show now in two different ways and contexts .
* fibration @xmath319 : * we use the notation of section [ sub sub : step 5 ]
. moreover we fix the following notation : @xmath363 and @xmath364 , @xmath365 , @xmath366 are respectively the non trivial components of the fibers of type @xmath78 , @xmath367 and @xmath367 respectively .
the class @xmath368 is the class of a section of infinite order of the fibration , generating the free part of @xmath25 and the class @xmath369 is the class of the 2-torsion section of the fibration .
the section @xmath152 meets the components @xmath370 , @xmath371 , @xmath372 and @xmath153 meets the components @xmath373 , @xmath374 , @xmath375 .
we observe that @xmath376 and @xmath377 which agree with ( * ? ? ?
* formula 22 ) and the fact that @xmath153 is a torsion section respectively .
we also give an explicit equation of this fibration and of its sections , see .
* fibration @xmath326 : * we use the notation of section [ sub sub : step 5 ] .
moreover we fix the following notation : @xmath378 and @xmath364 , @xmath365 , @xmath366 are respectively the non trivial components of the fibers of type @xmath78 , @xmath367 and @xmath367 respectively .
the class @xmath369 is the class of the 2-torsion section of the fibration .
observe that @xmath153 meets the components @xmath373 , @xmath374 , @xmath375 .
the class @xmath379 is the class of a section of infinite order , which intersects the following components of the reducible fibers : @xmath370 , @xmath374 , @xmath380 .
this agrees with the height formula .
we also give an explicit equation of this fibration and of its sections , see .
the generators of the free part of the mordell weil group is clearly defined up to the sum by a torsion section .
the section @xmath381 intersects the reducible fibers in the following components @xmath370 , @xmath382 , @xmath375 ( this follows by the group law on the fibers of type @xmath78 ( or @xmath79 ) and @xmath367 ) . comparing the sections of infinite order of the fibration 22 and the one of the fibration 22(b ) , one
immediately checks that their intersection properties are not the same , so the frames of the elliptic fibration 22 and of elliptic fibration 22(b ) are not the same and hence these two elliptic fibrations are in fact different under the @xmath58-classification .
we observe that both the fibrations @xmath383 and @xmath362 specialize the same fibration , which is given in ( * ? ? ?
* section 8.1 , table @xmath384 , case 11 ) ) .
indeed the torsion part of the mordell weil group , which is already present in the more general fibrations analyzed in @xcite , are the same and the difference between the fibration 22 and the fibration 22(b ) is in the free part of the mordell weil group , so the difference between these two fibrations involve exactly the classes that correspond to our specialization .
here we also give an equation for each of the two different fibrations # 22 and # 22(b ) .
both these equations are obtained from the equation of the elliptic fibration # 8 .
so first we deduce an equation for # 8 : let @xmath385 . substituting @xmath386 by @xmath387 in , we obtain the equation of an elliptic curve depending on @xmath388 , which corresponds to the fibration # 8 and with the following weierstrass equation @xmath389 * fibration # 22 * putting @xmath390 @xmath391 in we obtain @xmath392 with singular fibers of type @xmath393 we notice the point @xmath394 of height @xmath395 , therefore @xmath152 and @xmath396 generate the mordell - weil group of @xmath397 . to study the singular fiber at @xmath398 we do the transformation @xmath399 and @xmath400 with @xmath401 and @xmath402 we deduce that the section @xmath152 intersects the component of singular fibers at @xmath3 and @xmath403 with the same subscript , so this fibration corresponds to fibration # 22 . *
fibration # 22(b ) * putting @xmath404 in we obtain @xmath405 with singular fibers of type @xmath393 we notice the point @xmath406 of height @xmath395 , therefore @xmath152 and @xmath396 generate the mordell - weil group of @xmath407 .
since @xmath152 does not meet the node of the weierstrass model at @xmath408 , the section @xmath152 intersects the component @xmath409 of the singular fiber for @xmath410 so this fibration corresponds to # 22(b ) .
let us denote by @xmath411 and @xmath412 the elliptic fibrations # 9 and # 21 respectively .
they satisfy @xmath413 and @xmath414 , but @xmath411 is not @xmath58-equivalent to @xmath412 since , as above , the infinite order sections of these two fibrations have different intersection properties with the singular fibers .
indeed these two fibrations correspond to different fibrations ( * ? ? ?
* case 10a ) and case 10b ) , section 8.1 , table @xmath384 ) on the more general family of k3 surfaces considered in @xcite .
atkin , f. morain , _ finding suitable curves for the elliptic curve method of factorization .
comput . * 60 * , ( 1993 ) 399405 .
a. beauville , _ les familles stables de courbes elliptiques sur @xmath415 admettant quatre
fibres singulires . _ , c. r. acad .
paris sr .
* 294 * ( 1982 ) , 657660 .
bertin , _ mesure de mahler et srie @xmath90 dune surface k3 singulire_. actes de la confrence : fonctions @xmath90 et arithmtique .
actes de la confrence algbre thorie nbr . , lab .
besanon , ( 2010 ) 528 .
bertin , o. lecacheux _
elliptic fibrations on the modular surface associated to @xmath416 _ , in arithmetic and geometry of k3 surfaces and calabi - yau threefolds , 153199 , fields inst .
commun . , * 67 * , springer , new york , 2013 .
a. p. braun , y. kimura , t. watari , _ on the classification of elliptic fibrations modulo isomorphism on k3 surfaces with large picard number _ ,
j. w. s cassels , _ lectures on elliptic curves _ london math .
texts , * 24 * cambridge univ . press , cambridge ( 1991 ) .
p. comparin , a. garbagnati _ van geemen - sarti involutions and elliptic fibrations on k3 surfaces double cover of @xmath417 _ , journal of mathematical society of japan * 66 * ( 2014 ) 479522 .
d. cox , s. katz .
_ mirror symmetry and algebraic geometry_. providence : ams ( 1999 ) .
n. d. elkies ` http://www.math.harvard.edu/~elkies/k3_20si.html#-7[1 ` n. d. elkies _ three lectures on elliptic surfaces and curves of high rank , _ lecture notes , oberwolfach , 2007 .
n. elkies , m. schtt , _ genus 1 fibrations on the supersingular k3 surface in characteristic 2 with artin invariant 1 _ to appear in asian j. math .
a. garbagnati , a. sarti , _ elliptic fibrations and symplectic automorphisms on k3 surfaces _ , comm . in algebra
* 37 * ( 2009 ) , 36013631 .
d. karp , j. lewis , d. moore , d. skjorshammer , u. whitcher , _ on a family of k3 surfaces with @xmath160 symmetry_. arithmetic and geometry of k3 surfaces and calabi - yau threefolds , fields inst .
comm . , 2013 .
r. kloosterman , _ classification of all jacobian elliptic fibrations on certain k3 surfaces_. j. math .
japan * 58 * ( 2006 ) , 665680 . s. kondo , _ automorphisms of algebraic k3 surfaces which act trivially on picard groups _ , j. math .
japan , * 44 * ( 1992 ) , 7598 . s. kondo , _ algebraic k3 surfaces with finite automorphism group _ , nagoya math
j. volume * 116 * ( 1989 ) , 115 .
d. s. kubert , _
universal bounds on the torsion of elliptic curves _ , proc .
london math .
( 3 ) * 33 ( 1976 ) , 193237 .
a. kumar , _ elliptic fibrations on a generic jacobian kummer surface _ , arxiv:1105.1715 , to appear in j. algebraic geom .
m. kreuzer , h. skarke , _ classification of reflexive polyhedra in three dimensions_. adv .
phys . , * 2 * ( 1998 ) 853871 . m. kreuzer , h. skarke , _ complete classification of reflexive polyhedra in four dimensions_. adv .
phys . * 4 * ( 2000 ) 12091230 .
p. l. montgomery , _ speeding the pollard and elliptic curve methods of factorization _ math . comput . * 48 * , ( 1987 ) 243264 .
k. oguiso , _ on jacobian fibrations on the kummer surfaces of the product of nonisogenous elliptic curves _ , j. math .
japan * 41 * ( 1989 ) 651680 .
j. martinet , _ perfect lattices in euclidean spaces _ vol.327 springer - verlag berlin and heidelberg . v.v .
nikulin , _ integral symmetric bilinear forms and some of their applications _ , math .
* 14 * ( 1980 ) , 103167 .
nikulin , _ finite groups of automorphisms of khlerian k3 surfaces . _
( russian ) trudy moskov . mat .
* 38 * ( 1979 ) , 75137 .
english translation : trans .
moscow math .
* 38 * ( 1980 ) , 71135 .
k. nishiyama , _ the jacobian fibrations on some k3 surfaces and their mordell
weil groups _ , japan . j. math .
( n.s . ) * 22 * ( 1996 ) 293347 .
stein et al.,_sage mathematics software ( version 6.1 ) _ , the sage development team , 2014 , http://www.sagemath.org . m. schtt , t. shioda , _ elliptic surfaces _ , algebraic geometry in east asia
seoul 2008 , adv .
pure math .
, * 60 * , math .
japan , tokyo , ( 2010 ) 51160 .
m. schtt , _ k3 surface with picard rank 20 over @xmath144 _ algebra and number theory * 4 * ( 2010 ) 335356 .
t. shioda , _ on the mordell
weil lattices _ , comm .
pauli * 39 * ( 1990 ) , 211240 .
h. sterk , _ finiteness results for algebraic k3 surfaces _ , math
. z. , * 180 * ( 1985 ) , 507513 .
i. shimada , d .- q .
zhang , _ classification of extremal elliptic k3 surfaces and fundamental groups of open k3 surfaces _ , nagoya math . j. * 161 * ( 2001 ) , 2354 .
h. verrill , _ root lattices and pencils of varieties _ , j. math .
kyoto univ .
* 36 * ( 1996 ) , 423446 . | we classify , up to automorphisms , the elliptic fibrations on the singular k3 surface @xmath0 whose transcendental lattice is isometric to @xmath1 . |
rna polymerase ( rnap ) is a molecular motor @xcite .
it moves on a stretch of dna , utilizing chemical energy input , while polymerizing a messenger rna ( mrna ) @xcite .
the sequence of monomeric subunits of the mrna is dictated by the corresponding sequence on the template dna .
this process of template - dictated polymerization of rna is usually referred to as _ transcription _
it comprises three stages , namely , initiation , elongation of the mrna and termination .
we first report analytical results on the characteristic properties of single rnap motors . in our approach @xcite , each rnap is represented by a hard rod while the dna track is modelled as a one - dimensional lattice whose sites represent a nucleotide , the monomeric subunits of the dna .
the mechano - chemistry of individual rnap motors is captured in this model by assigning @xmath0 distinct `` chemical '' states to each rnap and postulating the nature of the transitions between these states .
the dwell time of an rnap at successive monomers of the dna template is a random variable ; its distribution characterizes the stochastic nature of the movement of rnap motors .
we derive the _ exact _ analytical expression for the dwell - time distribution of the rnaps in this model .
we also report results on the collective movements of the rnaps .
often many rnaps move simultaneously on the same dna track ; because of superficial similarities with vehicular traffic @xcite , we refer to such collective movements of rnaps as rnap traffic @xcite .
our model of rnap traffic can be regarded as an extension of the totally asymmetric simple exclusion process ( tasep ) @xcite for hard rods where each rod can exist at a location in one of its @xmath0 possible chemical states .
the movement of an rnap on its dna track is coupled to the elongation of the mrna chain that it synthesizes .
naturally , the rate of its forward movement depends on the availability of the monomeric subunits of the mrna and the associated `` chemical '' transitions on the dominant pathway in its mechano - chemical cycle . because of the incorporation of the mechano - chemical cycles of individual rnap motors , the number of rate constants in this model is higher than that in a tasep for hard rods .
consequently , we plot the phase diagrams of our model not in a two - dimensionl plane ( as is customary for the tasep ) , but in a 3-dimensional space where the additional dimension corresponds to the concentration of the monomeric subunits of the mrna .
we take the dna template as a one dimensional lattice of length @xmath1 and each rnap is taken as a hard rod of length @xmath2 in units of the length of a nucleotide .
although an rnap covers @xmath2 nucleotides , its position is denoted by the _
nucleotide covered by it .
transcription initiation and termination steps are taken into account by the rate constants @xmath3 and @xmath4 , respectively . a hard rod , representing an mrna , attaches to the first site @xmath5 on the lattice with rate @xmath3 if the first @xmath6 sites are not covered by any other rnap at that instant of time .
similarly , an mrna bound to the rightmost site @xmath7 is released from the system , with rate @xmath4 .
we have assumed hard core steric interaction among the rnaps ; therefore , no site can be simultaneously covered by more than one rnap . at every lattice site @xmath8
, an rnap can exist in one of two possible chemical states : in one of these it is bound with a pyrophosphate ( which is one of the byproducts of rna elongation reaction and is denoted by the symbol @xmath9 ) , whereas no @xmath9 is bound to it in the other chemical state ( see fig.[fig - model ] ) . for plotting our results ,
we have used throughout this paper @xmath10,@xmath11 and @xmath12~\tilde{\omega}_{21}^f ~s^{-1}$ ] , where @xmath13 $ ] is concentration of nucleotide triphosphate monomers ( fuel for transcription elongation ) and @xmath14 .
for every rnap , the dwell time is measured by an imaginary `` stop watch '' which is reset to zero whenever the rnap reaches the chemical state @xmath15 , _ for the first time _ , after arriving at a new site ( say , @xmath16-th from the @xmath8-th ) .
let @xmath17 be the probability of finding a rnap in the chemical state @xmath18 at time @xmath19 .
the time evolution of the probabilities @xmath17 are given by @xmath20 @xmath21 there is a close formal similarity between the mechano - chemical cycle of an rnap in our model ( see fig.[fig - model ] ) and the catalytic cycle of an enzyme in the michaelis - menten scenario @xcite .
the states @xmath15 and @xmath22 in the former correspond to the states @xmath23 and @xmath24 in the latter where @xmath23 represents the free enzyme while @xmath24 represents the enzyme - substrate complex . following the steps of calculation used earlier by kuo et al .
@xcite for the kinetics of single - molecule enzymatic reactions , we obtain the dwell time distribution @xmath25 \label{eq - ftgen}\end{aligned}\ ] ] where @xmath26 @xmath27 \(a ) + release ( and , hence , @xmath28 ) fixed , and ( b ) two different values of @xmath28 , keeping the ntp concentration fixed . , title="fig : " ] + ( b ) + release ( and , hence , @xmath28 ) fixed , and ( b ) two different values of @xmath28 , keeping the ntp concentration fixed . , title="fig : " ] the dwell time distribution ( [ eq - ftgen ] ) is plotted in fig.[fig - ft ] .
depending on the magnitudes of the rate constants the peak of the distribution may appear at such a small @xmath19 that it may not possible to detect the existence of this maxium in a laboratory experiment .
in that case , the dwell time distribution would appear to be purely a single exponential @xcite .
it is worth pointing out that our model does not incorporate backtracking of rnap motors which have been observed in the _ in - vitro _
experiments @xcite .
it has been argued by some groups @xcite that short transcriptional pausing is distinct from the long pauses which arise from backtracking .
in contrast , some other groups @xcite claim that polymerase backtracking can account for both the short and long pauses .
thus , the the role of backtracking in the pause distribution remain controversial .
moreover , it has been demonstrated that a polymerase stalled by backtracking can be re - activated by the `` push '' of another closely following it from behind @xcite .
therefore , in the crowded molecular environment of intracellular space , the occurrence of backtracking may be far less frequent that those observed under _ in - vitro _ conditions .
our model , which does not allow backtracking , predicts a dwell time distribution which is qualitatively very similar to that of the short pauses provided the most probable dwell time is shorter than 1 s. from equation ( [ eq - ftgen ] ) we get the inverse mean dwell time @xmath29 } } \label{eq - avtgen}\ ] ] where @xmath30 and @xmath31 .
the form of the expression ( [ eq - avtgen ] ) is identical to the michaelis - menten formula for the average rate of an enzymatic reaction .
it describes the slowing down of the `` bare '' elongation progress of an rnap due to the ntp reaction cycle that it has to undergo .
the unit of velocity is @xmath32 .
the fluctuations of the dwell time can be computed from the second moment @xmath33 } { \left(\omega_{12}~\omega_{21}^f \right)^2 } \end{aligned}\ ] ] of the dwell time distribution .
we find the randomness parameter @xcite @xmath34 is plotted against ntp concentration for three values of the parameter @xmath28 .
, title="fig : " ] + note that , for a one - step poisson process @xmath35 , @xmath36 .
the randomness parameter @xmath37 , given by ( [ eq - ranpar ] ) , is plotted against the ntp concentration in fig.[fig - ranpar ] for three different values of @xmath28 . at sufficiently low ntp concentration
, @xmath37 is unity because ntp binding with the rnap is the rate - limiting step . as ntp concentration increases
, @xmath37 exhibits a nonmonotonic variation . at sufficiently high ntp concentration , pp@xmath38-release ( which occurs with the rate @xmath28 ) is the rate - limiting step and ,
therefore , @xmath37 is unity also in this limit .
this interpretation is consistent with the fact that the smaller is the magnitude of @xmath28 , the quicker is the crossover to the value @xmath36 as the ntp concentration is increased .
the randomness parameter yields the diffusion coefficient @xcite @xmath39\nonumber\\ \label{eqn - diffusion}\end{aligned}\ ] ] the expression ( [ eqn - diffusion ] ) is in agreement with the general expression for the effective diffusion constant of a molecular motor with unbranched mechano - chemical cycle which was first reported by fisher and kolomeisky @xcite .
now we will take into account the hard core steric interaction among the rnaps which are simultaneously moving on the same dna track
. equations ( [ eq - masterp1 ] and [ eq - masterp2 ] ) will be modified to @xmath40 @xmath41 where @xmath42 is conditional probability @xcite of finding site @xmath43 ( @xmath44 for backward motion ) vacant , given there is a particle at site @xmath8 . due to the steric interactions between rnap s their stationary flux
@xmath45 ( and hence the transcription rate ) is no longer limited solely by the initiation and release at the terminal sites of the template dna .
we calculate the resulting phase diagram utilizing the extremum current hypothesis ( ech ) @xcite .
the ech relates the flux in the system under open boundary conditions ( obc ) to that under periodic boundary conditions ( pbc ) with the same bulk dynamics . in this approach
, one imagines that initiation and termination sites are connected to two separate reservoirs where the number densities of particles are @xmath46 and @xmath47 respectively , and where the particles follow the same dynamics as in the bulk of the real physical system . then @xmath48 the actual rates @xmath3 and @xmath4 of initiation and termination of mrna polymerization
are incorporated by appropriate choice of @xmath46 and @xmath47 respectively . .
the surfaces @xmath49 and @xmath50 separate the mc phase from the hd and ld phases , respectively .
, title="fig : " ] + -@xmath4 plane for several values of @xmath51 .
the numbers on the phase boundary lines represent the value of @xmath51 .
the inclined lines have ld and hd above and below , respectively , while the mc phase lies in the upper right corner . ]
an expression for @xmath52 was reported by us in ref . @xcite . in the special case where the dominant pathway is that shown in fig .
[ fig - model ] ( @xmath53 = 0 for further calculation as @xmath54 ) , we have @xmath55 the number density @xmath56 that corresponds to the maximum flux is given by the expression @xmath57^{-1}\ ] ] by comparing ( [ eq : effj ] ) with the exact current - density relation of the usual tasep for extended particles of size @xmath2 @xcite , which have no internal states ( formally obtained by taking the limit @xmath58 in the present model ) , we predict that the stationary current ( i.e. the collective average rate of translation ) is reduced by the occurrence of the intermediate state 1 through which the rnaps have to pass . from ( [ ech ] )
one expects three phases , viz .
a maximal - current(mc ) phase with with bulk density @xmath56 , a low - density phase ( ld ) with bulk density @xmath46 , and a high - density phase ( hd ) with bulk density @xmath47 .
using arguments similar to those used in ref .
@xcite in a similar context , we get @xcite @xmath59 and @xmath60 except that the projections are on @xmath51-@xmath4 plane for several values of @xmath3 .
the inclined lines have ld and hd above and below , respectively .
each vertical line separates the ld phase on the left from the mc phase on its right . ]
the condition for the coexistence of the high density ( hd ) and low density ( ld ) phases is @xmath61 with @xmath62 .
using the expression ( [ eq : effj ] ) for @xmath45 in ( [ eq - ldhd ] ) we get @xmath63 substituting ( [ eq - rhom ] ) and ( [ eq - rhop ] ) into ( [ eq - rhomp ] ) , we get the equation for the plane of coexistence of ld and hd to be @xmath64 where @xmath65 } .
\label{eq : phasec}\end{aligned}\ ] ] in order to compare our result with the 2-d phase diagram of the tasep in the @xmath66-plane , we project 2-d cross sections of the 3-d phase diagram , for several different values of @xmath67 onto the @xmath66-plane .
the lines of coexistence of the ld and hd phases on this projected two - dimensional plane are curved , a similar curvature is also reported by antal and schtz @xcite .
this is in contrast to the straight coexistence line for ld and hd phases of tasep .
the bulk density of the system is guided by following equations : @xmath68 \biggl[\dfrac{\omega_{12}\omega_{21}^f}{\omega_{12}+\omega_{21}^f}\biggr]~~\mbox{low density}\\ \rho_{+ } & ~ \mbox{if}~\omega_{\beta } < f(\omega_{\alpha},\omega_{21}^f ) { \rm and } ~\omega_{\beta } < \biggl[\dfrac{1-\rho_*\ell}{1-\rho_*(\ell-1)}\biggr ] \biggl[\dfrac{\omega_{12}\omega_{21}^f}{\omega_{12}+\omega_{21}^f}\biggr]~~\mbox{high density } \\
\rho _ { * } & ~\mbox{if}~\omega_{\beta}>\biggl[\dfrac{1-\rho_*\ell}{1-\rho_*(\ell-1)}\biggr ] \biggl[\dfrac{\omega_{12}\omega_{21}^f}{\omega_{12}+\omega_{21}^f}\biggr ] { \rm and } ~\omega_{\alpha } > \biggl[\dfrac{\rho_*}{1-\rho_*(\ell-1)}\biggr ] \biggl[\dfrac{\omega_{12}\omega_{21}^f}{\omega_{12}+\omega_{21}^f}\biggr]~~\mbox{maximal current}.\end{array } \right . \nonumber\\\end{aligned}\ ] ] in fig .
[ fig:3d ] , we plot the 3d phase diagram . except that
the projections are on @xmath51-@xmath3 plane for several values of @xmath4 . here
the inclined lines have hd and ld , respectively , above and below .
each vertical line separates the hd phase on the left from the mc phase on its right . ] in general , a plane @xmath69=constant intersects the surfaces i , ii and iii thereby generating the phase transition lines between the ld , hd and mc phases in the @xmath3-@xmath4 plane .
we have projected several of these 2d phase diagrams , each for one constant value of @xmath69 in figure [ fig : xyplane ] . in the inset ,
we have shown the value of @xmath51 for different lines .
we have also projected several 2d phase diagrams in the @xmath70-@xmath4 plane and @xmath51-@xmath4 plane , respectively , in figures [ fig : yzplane ] and [ fig : xzplane ] .
in this paper we have reported the exact dwell time distribution for a simple 2-state model of rnap motors . from this distribution
we have also computed the average velocity and the fluctuations of position and dwell time of rnap s on the dna nucleotides .
these expressions are consistent with a general formula derived earlier by fisher and kolomeisky for a generic model of molecular motors with unbranched mechano - chemical cycles . taking into account the presence of steric interactions between different rnap moving along the same dna template we have plotted the full 3d phase diagram of a model for multiple rnap traffic .
this model is a biologically motivated extension of the tasep , the novel feature being the incorporation of the mechano - chemical cycle of the rnap into the dynamics of the transcription process .
this leads to a hopping process with a dwell time distribution that is not a simple exponential .
nevertheless , the phase diagram is demonstrated to follow the extremal - current hypothesis @xcite for driven diffusive systems . using mean field theory
we have computed the effective boundary densities that enter the ech from the reaction constants of our model .
we observe that the collective average rate of translation as given by the stationary rnap current ( [ eq : effj ] ) is reduced by the need of the rnap to go through the pyrophosphate bound state .
this is a prediction that is open to experimental test .
the 2d cross sections of this phase diagram have been compared and contrasted with the phase diagram for the tasep . unlike in the tasep , the coexistence line between low- and high - density phase is curved for all parameter values .
this is a signature of broken particle - vacancy symmetry of the rnap dynamics .
the presence of this coexistence line suggests the occurrence of rnap `` traffic jams '' that our model predicts to appear when stationary initiation and release of rnap at the terminal sites of the dna track are able to balance each other .
this traffic jam would perform an unbiased random motion , as argued earlier on general theoretical grounds in the context of protein synthesis by ribosomes from mrna templates @xcite . + * acknowledgments * : this work is supported by a grant from csir ( india ) .
gms thanks iit kanpur for kind hospitality and dfg for partial financial support .
50 m. schliwa , ( ed . ) _ molecular motors _ , ( wiley - vch , 2003 ) .
j. gelles and r. landick , cell , * 93 * , 13 ( 1998 ) .
b. alberts et al .
_ essential cell biology _ , 2nd ed .
( garland science , taylor and francis , 2004 ) .
t. tripathi and d. chowdhury , phys .
e * 77 * , 011921 ( 2008 ) .
d. chowdhury , l. santen and a. schadschneider , phys .
rep . * 329 * , 199 ( 2000 ) .
s. klumpp and r. lipowsky , j. stat .
phys . * 113 * , 233 ( 2003 ) .
d. chowdhury , a. schadschneider and k. nishinari , phys . of life rev .
* 2 * , 318 ( 2005 ) .
m. voliotis , n. cohen , c. molina - paris and t.b .
liverpool , biophys .
j. * 94 * , 334 ( 2007 ) .
s. klumpp and t. hwa , pnas * 105 * , 18159 ( 2008 ) .
g. m. schtz , in : _ phase transitions and critical phenomena _ , vol .
19 ( acad . press , 2001 ) .
m. dixon and e.c .
webb , _ enzymes _ ( academic press , 1979 ) .
kou , b.j .
cherayil , w. min , b.p .
english and x.s .
xie , j. phys .
b * 109 * , 19068 - 19081 ( 2005 ) .
k. adelman , a. la porta , t.j .
santangelo , j.t .
lis , j.w .
roberts and m.d .
wang , pnas * 99 * , 13538 ( 2002 ) .
a. shundrovsky , t.j .
santangelo , j.w .
roberts and m.d .
wang , biophys .
j. * 87 * , 3945 ( 2004 ) .
abbondanzieri , w.j .
greenleaf , j.w .
shaevitz , r. landick and s.m .
block , nature * 438 * , 460 ( 2005 ) .
shaevitz , e.a .
abbondanzieri , r. landick and s.m .
block , nature * 426 * , 684 ( 2003 ) .
e. galburt , s.w .
grill , a. wiedmann , l. lubhowska , j. choy , e. nogales , m. kashlev and c. bustamante , nature * 446 * , 820 ( 2007 ) .
neuman , e.a .
abbondanzieri , r. landick , j. gelles and s.m .
block , cell * 115 * , 437 ( 2003 ) .
m. depken , e. galburt and s.w .
grill , biophys .
j. * 96 * , 2189 ( 2009 ) .
v. epshtein and e. nudler , science * 300 * , 801 ( 2003 ) .
m. schnitzer and s. block , cold spring harbor symp .
biol . * 60 * , 793 ( 1995 ) t. tripathi , ph.d .
thesis , iit kanpur ( 2009 ) .
fisher and a.b .
kolomeisky , pnas * 96 * , 6597 ( 1999 ) .
v. popkov and g. m. schtz , europhys . lett . * 48 * , 257 ( 1999 ) .
j. krug , phys . rev
. lett . * 67 * , 1882 ( 1991 ) . c. macdonald , j. gibbs and a. pipkin , biopolymers , * 6 * , 1 ( 1968 )
shaw , r.k.p .
zia and k.h .
lee , phys .
e * 68 * , 021910 ( 2003 ) .
g. schnherr and g.m .
schtz , j. phys .
a * 37 * , 8215 ( 2004 ) .
a. basu and d. chowdhury , phys .
e * 75 * , 021902 ( 2007 ) . t. antal and g. m. schtz , phys .
e * 62 * , 83 ( 2000 ) .
schtz , int . j. mod
b * 11 * , 197 ( 1997 ) . | polymerization of rna from a template dna is carried out by a molecular machine called rna polymerase ( rnap ) .
it also uses the template as a track on which it moves as a motor utilizing chemical energy input .
the time it spends at each successive monomer of dna is random ; we derive the exact distribution of these `` dwell times '' in our model .
the inverse of the mean dwell time satisfies a michaelis - menten - like equation and is also consistent with a general formula derived earlier by fisher and kolomeisky for molecular motors with unbranched mechano - chemical cycles .
often many rnap motors move simultaneously on the same track . incorporating the steric interactions among the rnaps in our model
, we also plot the three - dimensional phase diagram of our model for rnap traffic using an extremum current hypothesis . |
.volume fraction for @xmath98 , rlp . [ cols="^,^,^,^,^,^,^,^,^,^,^",options="header " , ]
according to the theory , the average voronoi volume for a packing with a distribution of radius @xmath109 , is given by the following self - consistent equation : where the different quantities are calculated as follow : @xmath111 @xmath112 to simplify we denoted @xmath113 , @xmath114 and @xmath80 the step - function .
@xmath115 @xmath116 @xmath117 @xmath118 @xmath119 @xmath120 | we develop a model to describe the properties of random assemblies of polydisperse hard spheres .
we show that the key features to describe the system are _ ( i ) _ the dependence between the free volume of a sphere and the various coordination numbers between the species , and _
( ii ) _ the dependence of the coordination numbers with the concentration of species ; quantities that are calculated analytically .
the model predicts the density of random close packing and random loose packing of polydisperse systems for a given distribution of ball size and describes packings for any interparticle friction coefficient .
the formalism allows to determine the optimal packing over different distributions and may help to treat packing problems of non - spherical particles which are notoriously difficult to solve .
understanding the basic properties of spheres packings is a major challenge since this problem may provide valuable knowledge regarding low temperature phases in condensed matter physics @xcite .
the canonical example is perhaps the monodisperse sphere packing problem .
it has been mathematically proven that the optimum way to arrange monodisperse spheres is the face - centered cubic lattice ; a problem that has been solved recently by hales , @xmath0 400 years after the famous kepler conjecture on the issue . on the other hand , it is commonly observed that packings arrange in a random fashion at a lower density state called random close packing or rcp @xcite .
furthermore , packings are mechanically stable up to an even lower limit called random loose packing , rlp . in parallel with the large literature dealing with monodisperse sphere packings , a large body of experimental
, theoretical and numerical work has been devoted to the analysis of polydisperse systems ; the interest arising due to the simple fact that polydispersivity is omnipresent in most realistic systems and industrial applications @xcite . while previous approaches have focused on frictionless packings , an integrated analytical approach that brings together different observations for all packings from rlp to rcp and for any friction or coordination number is still lacking .
based on our previous statistical mechanics approach @xcite , here we build such a framework .
we show that the key aspect to solve this problem is the dependence of the various coordination numbers between the different species and the concentration of the species .
this is calculated here and shown to agree well with computer simulations .
this result is then incorporated into a statistical theory of volume fluctuations as in @xcite which calculates the free volume of a particle in terms of the coordination number .
the main result is the prediction of the rlp and rcp limiting densities for a given distribution of ball sizes as well as the prediction of densities for any packing in between those limits .
the formalism allows for a determination of the best packing fraction in terms of different distribution of ball sizes with specified constraints , as we show with a simple example .
we discuss possible generalization of the method to solve more difficult problems like the phase behavior of systems of non - spherical particles like rods or spherocylinders in any dimensions ; problems of long - standing history in condensed matter @xcite .
recent theoretical advances @xcite allow for the prediction of the density of rcp and rlp for equal - size ball packings using a relation between the average volume and the geometrical coordination number . following this approach ,
we here describe long - range spatial correlations through a mean - field background term .
this approximation makes the problem amenable to analytic calculations , and is shown to describe well our simulation results .
an explicit inclusion of such correlations is possible in our framework , but severely complicates any solution attempts .
thus , we believe that the present approach is accurate enough for many important properties , such as the volume fraction calculation . the above theoretical framework will guide the present formalism for polydisperse systems .
we first treat the case of binary mixtures of hard spheres of radius @xmath1 and @xmath2 in 3d and then generalize the problem to any distribution and dimension .
* calculation of @xmath3. * the key quantity to calculate is @xmath3 , denoting the mean number of contacts of a ball of radius @xmath4 with a ball of radius @xmath5 , versus the concentration of one of the species . we need a formula for @xmath3 as a function of @xmath6 , the later being the size ratio , @xmath7 is the fraction of small balls in the packing @xmath8 with @xmath9 the number of @xmath10 balls and @xmath11 is the global geometrical coordination number averaged over all the particles : @xmath12 where @xmath13 is the average coordination of each species .
the coordinations are determined by three equations : @xmath14 we assume that these coordinations are inversely proportional to the average solid angle extended by contacting balls @xmath1 and @xmath15 .
the average solid angles are denoted @xmath16 and are calculated in terms of the solid angle that a ball @xmath5 occupy on a ball @xmath4 according to @xmath17 with ( see fig .
[ voro]a ) @xmath18 thus , @xmath19 represents the mean occupied surface on a @xmath10 ball weighted by the concentrations @xmath7 and @xmath20 .
this represents an approximation since the real weights are @xmath21 and @xmath22 , respectively . then , @xmath23 leading to the following normalizations : @xmath24 thus , the system of eqs .
( [ z1221 ] ) is reduced to a system of three equations for four unknowns @xmath3 . to close the system
we assume proportional laws and deduce @xmath3 from @xmath13 by considering that @xmath3 is proportional to the number of contacts of the @xmath10 balls times the number of contacts of the @xmath25 balls : @xmath26 using the first equation in ( [ z1221 ] ) we find the constants @xmath27 and @xmath28 , leading to the solution : @xmath29 figure [ fig : zandp]a , compares this solution to numerical simulations of hertz packings jammed at rcp @xcite for @xmath30 and @xmath31 .
we find that the formulae are very accurate for size ratios below 1.5 and present small deviations up to size ratio 2 .
the results are also in agreement with @xcite . *
the voronoi cell. * the common way to divide a system into volumes associated with each particle is the voronoi tessellation . the voronoi cell for monodisperse particles @xcite is composed by all the points nearest to the center of the ball than to any other ball .
this definition has been extended in @xcite to the case of polydisperse systems and non - spherical particles : instead of considering the classical voronoi polyhedron defined by the center of the particle , one should consider all the points which are closer to the surface of a given particle .
such a construction is called the _
voronoi s region _ and tiles a system of nonspherical convex particles and polydisperse systems as can be seen in fig .
[ voro]b . following this approach
we calculate the average volume of a polydisperse voronoi cell , denoted @xmath32 .
the volume fraction is given by @xmath33 , where @xmath34 is the mean volume of a ball .
we first find the analytical formula of the voronoi s region .
the boundary of the voronoi cell in the direction @xmath35 of a @xmath10 ball next to a @xmath25 ball at position @xmath36 is ( fig .
[ voro]b ) : @xmath37 where @xmath35 and @xmath38 are unitary .
thus , the boundary of a voronoi cell of a ball @xmath10 in the direction @xmath35 is the minimum of @xmath39 over all the particles @xmath25 for any @xmath40 .
this leads to @xmath41 the volume of the cell of the ball @xmath10 is then given by @xmath42 .
we define the orientational voronoi volume , @xmath43 , along the direction @xmath35 by @xmath44 .
this leads to @xmath45 this definition leads to the results of @xcite when @xmath46 .
since the system is isotropic , the mean voronoi volume can be calculated as : @xmath47 * calculation of the mean voronoi volume. * having calculated the voronoi cell exactly in eq .
( [ voronoi ] ) , we now proceed to develop a probability theory of volume fluctuations in the spirit of the quasiparticle approximation used in @xcite to obtain the mean voronoi volume . for a given ball @xmath10
, the calculation of @xmath48 reduces to finding the ball @xmath49 that minimizes @xmath39 .
we call @xmath49 the voronoi ball for the ball @xmath10 .
we consider @xmath50 , @xmath51 and @xmath52 .
we have @xmath53 .
therefore , we just need to compute the inverse cumulative distribution function , denoted @xmath54 , to find all the balls @xmath25 with @xmath55 , and thus not contributing to the voronoi volume of the ball @xmath10 .
the average voronoi volume is then given by the expression @xmath56 we calculate the mean voronoi volume for the balls of radius @xmath1 and @xmath15 separately and then average them .
we denote @xmath57 and @xmath58 the inverse cumulative distributions respectively , and @xmath59 and therefore , @xmath60 * calculation of @xmath54. * there are three salient steps in the calculation of @xmath54 : _ ( i ) _ the separation of @xmath54 following eq .
( [ p_c ] ) . _
( ii ) _ the separation of each term @xmath61 , @xmath62 , into two contributions : a term taking into account the contact spheres , @xmath63 , and a bulk term , @xmath64 .
the contact term clearly depends on @xmath3 .
the bulk term averages over all spatial correlations of non - contact particles and , without significant loss of accuracy as shown below , we assume that it only depends on the average value of @xmath32 . in principle , it is possible to use a more realistic form for this term , but this would render the problem practically unsolvable . _ ( iii ) _ the separation of @xmath63 and @xmath65 into two terms @xmath66 and @xmath67 , @xmath68 , for each species .
an assumption of the theory ( to be tested a posteriori with computer simulations ) is that all of these terms are independent .
thus @xmath69 also , we work in the limit of large number of particles leading to boltzmann - like exponential distributions for each @xmath70 and @xmath71 @xcite .
four important quantities are then defined : _ ( i ) _ @xmath72 and _ ( ii ) _ @xmath73 : the excluded volume and surface on the ball , respectively , where no center of a ball @xmath25 can be located for a given ball @xmath10 and for a given @xmath74 . _ ( iii ) _ @xmath75 : the mean number of balls @xmath25 by the left free volume . _ ( iv ) _ @xmath76 : the mean number of balls @xmath25 by the left free surface on a ball @xmath10 .
these considerations lead to : @xmath77 next , we calculate these four quantities . to simplify we denote @xmath78 , @xmath79 and @xmath80 the step - function .
we obtain : @xmath81 where @xmath82 and @xmath83 .
the fourth quantity , @xmath76 , is the most difficult to calculate . in terms of the occupied areas
eqs .
( [ occ ] ) we have @xmath84
. however , for the contact terms , the analogy with the boltzmann - like exponential distribution of volumes is far from being exact .
this is because this form is based on the large number limit which in the case of contacting balls is no more than around 6 .
therefore , the exponential distribution is used as an ansatz with @xmath76 a variational parameter as in @xcite .
we denote @xmath85 the mean solid angle of the gaps left between the @xmath25 contacting neighbors of a @xmath10 ball ( fig .
[ voro]a ) .
we obtain : @xmath86 then , we perform numerical simulations to find @xmath87 .
we randomly generate balls of radius @xmath4 and @xmath5 with the proportion @xmath21 and @xmath22 respectively around a ball of radius @xmath4 and then evaluate the mean free surface @xmath88 and its inverse to obtain @xmath76 .
we find @xmath89 which is a generalization of the results of @xcite to polydisperse systems .
using eqs .
( [ pij ] ) , ( [ sstar ] ) and ( [ rhos ] ) into ( [ pc ] ) , @xmath54 can be calculated by solving the equations numerically .
figure [ fig : zandp]b depicts the comparison of the theoretical results of the probability of voronoi volumes @xmath54 with computer generated hertzian packings with @xmath31 for @xmath90 at rcp .
the calculated distribution is quite accurate for most of the range except for small deviations at high values of @xmath74 , which however , do not affect much the value of the average @xmath91 in eq .
( [ ww ] ) .
this shows that our mean - field approximation already captures the main contribution of the probability distribution function @xmath54 .
* calculation of @xmath32. * the above considerations lead to a final form to calculate @xmath92 using eq .
( [ pc ] ) into ( [ ww ] ) : @xmath93 notice that @xmath94 depends on @xmath32 , eq .
( [ sstar ] ) , and @xmath95 depends on the @xmath3 , eq .
( [ rhos ] ) , which in turn depends on the concentration @xmath7 and @xmath11 through eq .
( [ zs ] )
. therefore eq .
( [ final ] ) is a self - consistent equation to obtain @xmath96 , for a given @xmath97 . a numerical integration of eq .
( [ final ] ) is performed versus @xmath7 for a given value of @xmath11 . by considering the isostatic limits of @xmath31 and @xmath98 we predict the limits of rcp and rlp at zero friction and infinite friction between the spheres , respectively @xcite .
the results for the volume fraction at rcp versus @xmath7 are depicted in fig .
[ fig : jamcompare]a which also compares the results to numerically generated packings of hertz spheres @xcite .
we see a very good agreement indicating that the theory captures well the behavior of polydisperse packings and that the approximations used are reasonable . for size ratios larger than 2 deviations
are found indicating the limit of validity of the approach .
for any other value of interparticle friction between 0 and @xmath99 , the density is obtained by setting @xmath11 between the limiting isostatic values of 6 and 4 , respectively .
the resulting volume fraction is shown in fig .
[ fig : jamcompare]b .
the result for rlp for a given @xmath7 is a new prediction as this problem has not been investigated before .
our results promote new experiments to test the rlp limit of polydisperse systems shown in fig .
[ fig : jamcompare]b . the formalism can be extended to consider any distribution of sphere radius .
the main modification is that all the sums over the radius are replaced by integrations over the desired distribution of radius @xmath100 ( the binary case corresponds to two delta - functions at @xmath1 and @xmath15 ) . and
all the quantities are calculated for balls of internal radius @xmath101 and external @xmath102 and @xmath7 and @xmath20 are replaced by @xmath103 and @xmath104 , respectively , including the coordinations ( see supplementary information b ) .
this result allows to explore the space of radius distributions in search of the optimal packings for a given polydispersivity .
this analysis could be of industrial interest if one wishes to optimize the packing fraction by introducing different species in the mixture .
\(a ) versus @xmath7 for different values of @xmath97 as indicated .
error bars are std over 10 realizations of the packings with 10,000 balls .
( b ) three dimensional surface plot of @xmath105 as a function of @xmath11 and @xmath7 for @xmath106 .
the numerical results at rcp and rlp are provided in supplementary information a.,title="fig:",scaledwidth=25.0% ] ( b ) versus @xmath7 for different values of @xmath97 as indicated .
error bars are std over 10 realizations of the packings with 10,000 balls .
( b ) three dimensional surface plot of @xmath105 as a function of @xmath11 and @xmath7 for @xmath106 .
the numerical results at rcp and rlp are provided in supplementary information a.,title="fig:",scaledwidth=25.0% ] we calculate the volume fraction for several distributions @xmath100 constraint by ball radius in the range [ 1,2 ] , in search of the optimal packing .
we calculate the volume fraction for various @xmath100 ranging from uniform to two - peaked gaussian distributions of varied widths .
we find that the more small balls we add the better the packing until a certain point where the volume fraction starts to decrease .
this maximum can be rationalized assuming that the small balls always fill the gaps between the large ones as long as there are enough large balls .
further extensions of the theory to any dimension can be performed by replacing @xmath107 by d in eq .
( [ voronoi ] ) and developing a theory of volume fluctuation in d - dimensions .
we notice that many of the approximations employed in 3d may become exact for large d , thus we expect improved results in the mean field limit of infinite dimensions .
the method allows to treat more difficult problems .
for instance , the prediction of the volume fraction of a jammed system of non - spherical particles is a long - standing problem .
theoretical predictions of onsager @xcite are valid for large aspect ratios , like elongated rods .
experiments however , find interesting new physics for small aspect ratios . in this respect
, the present polydisperse theory could be mapped to the problem of ellipsoids , spherocylinders or rods .
a voronoi cell needs to be calculated as a function of the angles defining the orientation of the non - spherical particles in analogy of the calculation between two particles of different radii .
the integration over @xmath108 in eq .
( [ final ] ) is then replaced by integration over weighted orientational angles . the above analysis can also be extended to dimensions beyond three @xcite .
although many of the appproximations should work better in higher dimensions , some of the hypotheses ( for example , the contact term ansatz ) need to be reassessed .
thus , higher dimensions studies can not be addressed as trivial extensions and need to be handled with care . in summary
, a theoretical framework is presented that predicts the rlp and rcp limits of a system of polydisperse spheres and brings together distinct results into a common theoretical framework .
the formalism has the potential to solve other problems in condensed matter physics such as the mixing and phase behavior of systems of hard particles of different shapes and size . 99 a. coniglio , a. fierro , h. j. herrmann , m. nicodemi , eds , _ unifying concepts in granular media and glasses _
( elsevier , amsterdam , 2004 ) .
j. d. bernal , j. mason , nature * 188 * , 910 ( 1960 ) . j. dodds , nature * 256 * , 187 ( 1975 ) ; k. de lange kristiansen , a. wouterse , a. philipse , physica a * 358 * , 249 ( 2005 ) ; m. clusel , e. i. corwin , a. o. n. siemens , j. bruji , nature * 460 * , 611 ( 2009 ) .
i. biazzo , f. caltagirone , g. parisi , f. zamponi , phys .
rev .
lett . * 102 * , 195701 ( 2009 ) .
c. song , p. wang , h. a. makse , nature * 453 * , 629 ( 2008 ) ; c. briscoe , c. song , p. wang , h. a. makse , phys .
rev .
lett .
* 101 * , 188001 ( 2008 ) .
l. onsager , ann . n. y. acad
. sci . * 51 * , 627 ( 1949 ) .
v. a. luchnikov , n. n. medvedev , l. oger , j .-
p .
troadec , phys .
rev .
e * 59 * , 7205 ( 1999 ) .
j. a. van meel , b. charbonneau , a. fortini , p. charbonneau phys .
rev .
e , 80,061110 , ( 2009 ) . |
we shall be concerned here with the current - current correlators in the coordinate space : [ twopoint ] ( x)= 0|j(x)j^(0)|0 , in case of the @xmath1 and ( pseudo)scalar currents : j_^(va)&=&|q_i_(1_5)q_j , + j^sp&=&[(m_i- m_j)(m_i+m_j)]|q_i(1_5)q_j , where @xmath2 and @xmath3 are the quark fields and masses .
the two - point functions ( [ twopoint ] ) obey a dispersion representation : ( x)=_0^dt k_1(x ) im(t ) , where im@xmath4 is related to the current induced cross section and @xmath5 is the modified bessel function , which behaves for small @xmath6 as : k(z0)+z . in the limit @xmath7
, @xmath8 coincides with the free - field correlator and the main theoretical issue is how the asymptotic freedom gets violated at intermediate @xmath0 .
+ from pure theoretical point of view , the use of the @xmath0-space is no better than the use of the momentum space , which is the traditional tool of the qcd sum rules @xcite .
each representation has its own advantages and inconveniences ( for a recent discussion see @xcite ) .
the @xmath0-space approach is motivated and described in detail in ref .
@xcite . in particular , the current correlators ( [ twopoint ] ) are measured in the most direct way on the lattice .
the importance of the lattice measurements @xcite is that they allow to study the correlation functions for currents with various quantum numbers , while direct experimental information is confined to only vector and axial - vector currents @xcite .
the well - known @xmath9-decay data were widely used for theoretical analyses both in the @xmath10- and @xmath0-spaces ( see , e.g. , @xcite ) .
most recently , new lattice data on the @xmath11 channels were obtained @xcite .
the most interesting observation is that in the @xmath12 channel there are noticeable deviations from the instanton liquid model @xcite while in the @xmath13 channels the agreement of the existing data with this model is quite good @xcite .
+ such deviations were in fact predicted in ref .
@xcite where unconventional quadratic corrections , @xmath14 were introduced .
the primary aim of the present note is to perform a more detailed comparison of the lattice data with the model of ref .
we , indeed , find further support for the quadratic corrections .
however , the overall picture is far from being complete and we are trying to analyze the data in a more generic way .
the central assumption is that the violations of the parton model for the correlators at moderate @xmath0 are due to power - like corrections .
for the sake of completeness , we begin with a summary of theoretical expressions for the current correlators , both in the @xmath15 and @xmath16spaces .
we will focus on the @xmath1 and @xmath17 channels since the recent lattice data @xcite refer to these channels . in case of @xmath1 currents
the correlator is defined as : [ odin ] _ ( q)=id^4x e^iqxtj_(x)j_(0)^= ( q_q_-g_q^2)(q^2 ) , where @xmath18 in the euclidean space - time . for the sake of definiteness we fix the flavor structure of the light - quark current @xmath19 as : j_^va=|u_(1_5 ) d . in the chiral limit one has in the @xmath20 case ( see ,
e.g. , @xcite ) : [ sum ] ^v+a(q^2 ) = 12 ^ 2 - 1+_s -_s^2q^2 + 3_s(g^a_)^2q^4 + 256 ^ 381_s|qq^2q^6 .
the corresponding relation for the @xmath21 case reads as : [ difference ] ^v - a(q^2 ) = - , in the x - space the same correlators , upon dividing by @xmath22 where @xmath22 stands for the perturbative correlator , are obtained by applying the equations collected for convenience in the table [ fourier ] .
.some useful fourier transforms [ cols= " < , < " , ] the analysis indicates that a much better fit of the lattice data for the @xmath12 channel at moderate values of @xmath0 is achieved after the inclusion of the @xmath23 , or @xmath24 quadratic correction .
a caveat is that we account only for the power corrections , not pure perturbative contributions .
the reason is that the lattice data , in their present status , do not give any clear indication of the perturbative contributions .
note also that the data can not discriminate between the values of the dimension four and six condensates entering in set 1 and set 2 as the effects of these two condensates tend to compensate each other for the choice @xmath25 .
the agreement of the ope with the lattice data at larger values of @xmath0 can be obtained by the inclusion of the @xmath26 condensate with a size @xmath27 where we have used @xmath28 .
this value can be compared with the one @xmath29 , which one would obtain from the evaluation of these contributions in @xcite and where a modified factorization of the gluon condensates proposed in @xcite has been used .
channel : comparison of the lattice data from @xcite with the ope predictions for the set 3 qcd condensates values given in table 2 including a fitted value of the @xmath26 contributions .
the diamond curve is the prediction from the instanton liquid model of @xcite .
, width=491 ] for completion , we show in fig .
2 , a fit of the lattice data in the v+a channel using set 3 values of the gluon and quark condensates and quadratic term plus a @xmath26 contribution with the strength @xmath30 to be compared with the one @xmath31 which one would obtain using the results in @xcite .
both fits in figs 1 and 2 might indicate that the vacuum saturation can be strongly violated for higher dimension condensates , a feature already encountered from different analysis of the @xmath9 and @xmath32 data @xcite .
therefore , we would also expect analogous large deviations in the v - a channel .
while evaluating the emerging picture of the power corrections , one should face the possibility that the standard ope ( see , e.g. , @xcite ) is valid only at very short distances .
what is even more important , the mass scale where higher terms in the ope become numerically comparable to the lowest ones is not necessarily the scale associated with the resonances but could be considerably higher .
there is accumulating evidence to support such a view : + ( 1 ) a direct comparison of the ope with the lattice data in the @xmath21 channel demonstrates that the convergence radius of the ope is no larger than 0.3 fm @xcite . + ( 2 ) within the instanton liquid model @xcite , the distance between instantons is a few times larger than the size of the instantons . on the physical grounds
, the ope applies at distances smaller than the instanton size while the resonance properties are rather related to distance between the instantons ( if encoded in the model at all ) . respectively ,
neither the lattice data nor the predictions of the instanton liquid model exhibit any irregularity at the convergence radius of the ope @xcite .
+ ( 3 ) within the monopole - dominated - vacuum model the two scales are even more pronounced numerically : the monopole radius is about 0.06 fm while the distance between the monopoles is about 0.5 fm @xcite . + ( 4 ) if one replaces the local condensates of the standard ope by their non - local counterparts ( for a review see @xcite ) , then the effect of the non - locality is strong already at @xmath33 fm @xcite .
if , indeed , the validity of the standard ope derived within the fundamental qcd is shrunk to very short distances , then improving within the standard ope fits to the data obtained at presently available lattices might not be a proper criterion for selecting the right model .
instead , there emerges a picture according to which the non - standard quadratic power corrections dominate the presently available `` intermediate distances '' of about @xmath34 fm .
there are two pieces of evidence of the quadratic corrections dominating over the whole range of intermediate distances indicated above : + ( 1 ) non - perturbative contribution to the heavy quark potential is linear at all the distances @xmath35 fm ( for discussion see @xcite ) .
+ ( 2 ) instanton density , as function of the instanton size @xmath36 is reproduced at all the distances @xmath37 fm @xcite .
+ moreover , + ( 3 ) introduction of the tachyonic gluon mass explains in a simple and unified way existence of the strong channel dependence of the scales of the violation of the asymptotic freedom @xcite .
+ to this list we add now a new observation : + ( 4 ) instanton - liquid model plus the @xmath38 correction gives a reasonable fit in the @xmath39 channel at distances @xmath34 fm .
as for the status of the @xmath38 correction it is most solid within the effective higgs - like theories which are common within the monopole mechanism of confinement .
indeed , in the presence of a magnetically charged ( effective ) scalar field the symmetry of the theory is @xmath40 @xcite . upon the spontaneous breaking of the magnetic @xmath41 the gauge boson acquires a non - vanishing mass and
its mass squared is the only parameter of dimension @xmath42 consistent with the symmetry .
moreover , in exchanges between ( color ) charged particles the gauge - boson mass appears to be tachyonic mass as was demonstrated on the @xmath41 example in refs .
detailed analysis of various power corrections within the higgs - like models can be found in refs .
moreover , if the monopole size is indeed as small as indicated above , then the effective higgs - like theories can apply at all the distances @xmath43 fm . to summarize
, the @xmath38 corrections introduced in @xcite rather drastically improve agreement of the theoretical predictions for the current correlator in the @xmath39 channel with the lattice data ( without affecting the other channels measured ) , in the moderate @xmath0-region , and which can be extended to higher @xmath0-values by including higher dimension condensates .
the main uncertainty of the analysis is due to neglect of the pure perturbative corrections , not detected so far on the lattice .
further checks could be provided by measuring the current correlators in the other channels discussed in @xcite .
the success of the fits with @xmath38 corrections included can be understood within the effective theories .
the authors are grateful to th . de grand , f.v .
gubarev , e.v .
shuryak for useful discussions and communications .
s. narison wishes to thank w - y .
pauchy hwang for the hospitality at cospa - ntu ( taipei ) , where some part of this work has been done .
v. zakharov would like to acknowledge gratefully the hospitality of the centre national de la recherche scientifique ( cnrs ) during his stay at the university of montpellier when this was begun .
shifman , a.i .
vainshtein and v.i .
zakharov , _ nucl .
* 147 * ( 1979 ) 385 , 448 .
s. narison , _ qcd spectral sum rules , lecture notes in physics _
26 * ( 1989 ) ed .
world scientific .
v. shevchenko , yu.simonov , hep - ph/0109051 .
shuryak , _ rev .
* 65 * ( 1993 ) 1 ; + t. schafer , e.v .
shuryak , _ rev .
phys . _ * 70 * ( 1998 ) 323 .
chu , j.m .
grandy , s. huang , j.w .
negele , _ phys .
_ * d48 * ( 1993 ) 3340 .
t. degrand , hep - lat/0106001 .
the aleph collaboration , r. barate et al .
j. _ * c 76 * ( 1997 ) 15 ; * c 4 * ( 1998 ) 409 .
the opal collaboration , k. ackerstaff et al .
j. _ * c 7 * ( 1999 ) 571 .
e.braaten , s. narison and a. pich , _ nucl .
* b 373 * ( 1992 ) 581 .
donoghue and e. golowich , _ phys .
_ * b 478 * ( 2000 ) 172 ; + e. de rafael , _ nucl .
suppl . ) _ * b 96 * ( 2001 ) 316 ; + s. narison , _ nucl . phys . _
* b 593 * ( 2001 ) 3 ; _ nucl . phys .
( proc . suppl . )
_ * b 96 * ( 2001 ) 364 .
ioffe , k.n .
zyablyuk , _ nucl .
phys . _ * a 687 * ( 2001 ) 437 , + b.v .
geshkenbein , b.l .
ioffe , k.n .
zyablyuk , hep - ph/0104048 .
t. schafer and e.v .
shuryak , _ phys .
* 86 * ( 2001 ) 3973 ; hep - ph/9610451 .
chetyrkin , s. narison and v.i .
zakharov , _ nucl .
* b 550 * ( 1999 ) 353 . c. becchi , s. narison , e. de rafael and f.j .
yndurain , _ z. phys . _
* c 8 * ( 1981 ) 335 .
s. narison , _ phys .
* b 361 * ( 1995 ) 121 ; * b 387 * ( 1996 ) 162 .
pdg 2000 , d.e .
groom et al .
j. _ * c 15 * ( 2000 ) 1 .
s. narison , n. paver , d. treleani and e. de rafael , _ nucl .
_ * b 212 * ( 1983 ) 365 .
( 1993 ) 1 .
g. launer , s. narison and r. tarrach , _ z. phys . _
* c 26 * ( 1984 ) 209 .
s. narison , _ phys .
* b 300 * ( 1993 ) 293 .
dominguez and k. schilcher , _ phys .
* d 61 * ( 2000 ) 114020 .
d. broadhurst and s.c .
generalis , _ phys .
* b 165 * ( 1985 ) 175 .
e. bagan , j.i .
latorre , p. pascual , r. tarrach , _ nucl .
* b 254 * ( 1985 ) 555 .
bertlmann , c.a .
dominguez , g. launer , m. perrottet , e. de rafael , _ z. phys . _
* c 39 * ( 1988 ) 231 . v.g .
bornyakov , m.n .
chernodub , f.v .
gubarev , m.i .
polikarpov , t. suzuki , a.i .
veselov , v.i .
zakharov , hep - lat/0103032 .
dosch , v.i .
shevchenko , yu.a .
simonov , hep - ph/0007223 .
r. hofmann , hep - ph/0109009 , hep - ph/0109008 , hep - ph/0109007 .
gubarev , m.i .
polikarpov , v.i .
zakharov , _ mod .
lett . _ * a14 * ( 1999 ) 2039 ; + m.n .
chernodub , f.v .
gubarev , m.i .
polikarpov , v.i .
zakharov , _ phys .
* b475 * ( 2000 ) 303 ; + f.v .
gubarev , v.i .
zakharov , _ phys .
* b501 * ( 2001 ) 28 .
shuryak , hep - ph/9909458 .
r. akhoury , v.i .
zakharov , _ phys .
* b438 * ( 1998 ) 165 ; + f.j .
yndurain , _ nucl .
suppl ) _ * b64 * ( 1998 ) 433 ; + f.v .
gubarev , m.i .
polikarpov , v.i .
zakharov , hep - ph/9908292 .
chernodub , f.v .
gubarev , m.i .
polikarpov , v.i .
zakharov , _ nucl .
phys . _ * b592 * ( 2001 ) 107 ( hep - th/0003138 ) ; _ nucl
_ * b600 * ( 2001 ) 163 ( hep - th/0010265 ) . | we consider an interpretation of the recent lattice data on the current - current correlators in the @xmath0-space .
the data indicate rather striking difference between ( axial)vector and ( pseudo)scalar channels which goes beyond the predictions of the standard non - perturbative models .
we argue that if the difference is to be explained by power corrections , there is a unique choice of the form of the correction .
we discuss the emerging picture of the power corrections . |
in 1938 , kaptiza @xcite and allen / misener @xcite independently discovered that , below 2.17 k , liquid helium-4 can flow without friction .
this superfluidity is also a property of dilute ultracold atomic gases , and has been demonstrated through the presence of a critical velocity @xcite , the formation of arrays of vortices under rotation @xcite and the observation of persistent flow @xcite .
the high degree of control over both the gas interactions and the trapping potential allows detailed studies of superfluidity in these systems , whilst their accurate modelling is also possible using the mean - field gross - pitaevskii equation .
there have been numerous alternative proposals for toroidal atom traps using magnetic fields with @xcite or without rf - dressing @xcite , electrostatic fields @xcite , time - averaged optical fields @xcite or laguerre - gauss beams @xcite .
an advantage of using optical rather than magnetic potentials is that they are suitable for trapping atoms in any state . in order for any of these to be used to study superflow of ultracold atoms , a stirring mechanism needs to be introduced , such as the transfer of orbital angular momentum from an additional light source @xcite or a stirring laser beam @xcite .
we propose a new method of generating optical ring traps with which to study superfluidity phenomena .
the trapping potential is created by a computer generated hologram ( cgh ) and a phase - only spatial light modulator ( slm ) , and incorporates a controllable stirring mechanism using the same laser beam .
our slm is a programmable diffractive optical device which consists of a two - dimensional array of pixels .
each of these is individually controllable , imprinting an arbitrary phase onto incident light .
if the diffracted light is then focussed by a lens , the intensity distribution in the focal plane of the lens is given by the fourier transform of the phase pattern on the slm , allowing freedom to create arbitrary intensity distributions . by displaying a sequence of phase patterns on the slm
, dynamic light patterns can also be generated .
cghs are becoming increasingly popular as a method to trap ultracold atoms .
slms have been proposed as a tool for generating a number of exotic trapping geometries beyond those achievable with standing - wave technologies , such as atom interferometers @xcite and ring lattices @xcite , while they have been used to manipulate single atoms @xcite , clouds of cold atoms from a magneto - optical trap @xcite and a bose - einstein condensate @xcite .
all works to date have used either arrays of dipole potentials or patterns for which the phase is well known , e.g. laguerre - gauss beams . in general , calculating a phase pattern which will produce an arbitrary desired intensity distribution is not an easy task as the undefined phase in the fourier plane gives the problem many solutions .
however , the calculation can be performed using relatively slow direct search algorithms @xcite , computationally - demanding but highly - accurate genetic algorithms@xcite or computationally - efficient iterative fourier transform algorithms such as gerchberg - saxton @xcite or adaptive - additive @xcite . in 2008 , pasienski and demarco introduced a variant of these algorithms : the mixed - region amplitude freedom ( mraf ) algorithm @xcite .
this allows the creation of smooth , continuous two - dimensional optical traps .
however , until now experimental achievement of optical traps suitable for ultracold atoms using this algorithm has not been shown . by recording intensity patterns on a ccd camera , we show that the mraf algorithm can generate light patterns suitable for trapping a bose
einstein condensate .
we further demonstrate that this trap can be dynamically varied in order to induce superflow .
we also find that the physical implementation of these cghs using slms is susceptible to aberrations in the optical system and imperfect device response , thus introducing errors and roughness that were not present during the calculation .
however , we show that this roughness can be reduced by incorporating the mraf algorithm into a feedback loop .
finally , we discuss the future integration of our trap into a cold atoms experiment .
a conservative trapping potential for ultracold atoms can be generated with focussed laser light far - detuned from an atomic transition .
light of intensity @xmath0 which is detuned by @xmath1 from an atomic transition frequency @xmath2 gives a trapping potential @xmath3 where @xmath4 is the natural linewidth of the atomic transition and @xmath5 is the speed of light .
we design an annular light pattern of intensity @xmath6 with two bright spots of intensity @xmath7 , given by @xmath8 where @xmath9 is the radius of the ring , @xmath10 is the angular position of the bright spots , and @xmath11 is the @xmath12 waist of the gaussian ring potential , as shown in figure [ fig : targpred ] .
the mraf algorithm divides the output plane into three regions : the measure region , the signal region and the noise region .
the measure region closely matches the boundaries of the light pattern , whilst the signal region is large enough to contain the measure region plus a border which will be devoid of light .
the amplitude of the light in these regions is fixed to match the target amplitude , meaning `` noise '' on the pattern can only exist in the noise region ( the entire plane outside the signal region ) , where the amplitude is unconstrained .
the algorithm converges to a solution within 100 iterations and the rms error of the calculated output is @xmath13 . in the measure region
this has been achieved by allowing unwanted light to be present outside the signal region.,scaledwidth=50.0% ] we plan to induce superflow by `` stirring '' the atoms in the ring potential by rotating the bright spots within the pattern .
this is possible by replacing the phase pattern on the slm with one which corresponds to the next position of the bright spots , as shown in figure [ fig : rotred]a .
the experiments reported in @xcite showed superflow of atoms with angular momentum @xmath14 and @xmath15 , which for their trap corresponds to a rotational frequency of around @xmath16hz and @xmath17hz respectively
. the maximum rate at which our slm can refresh the whole pattern is @xmath18hz , which allows a maximum of @xmath19 steps within one complete rotation of our ring trap .
the difference between consecutive positions of the bright spots should thus be sufficiently small that the motion of the stirring appears continuous to the atoms .
once superflow has begun , the stirring beams can be gradually removed , as shown in figure [ fig : rotred]b . as in @xcite
, we plan to ramp the value @xmath20 to zero within @xmath21ms , which allows us to refresh the pattern ten times .
our slm ( boulder nonlinear systems p256 ) has @xmath22 nematic liquid crystal pixels , each of which can shift the phase of an incoming laser beam in steps of @xmath23 .
we send the phase pattern calculated using mraf to our slm , which is illuminated with collimated and spatially filtered light from a @xmath24 nm diode laser . to illuminate the slm as uniformly as possible ,
we expand the beam to a @xmath12 radius of @xmath25 mm to overfill the active area of the slm , a square of side length @xmath26 mm .
this ensures that all pixels of the slm are illuminated by a comparable light power .
mraf is capable of calculating the phase required to generate the desired intensity distribution regardless of the incoming beam profile , so in future the overfilling could be relaxed to allow more power in the optical trap .
the fact that mraf shifts light into the noise region coupled to losses due to our overfilling and to the generation of other diffracted orders gives us @xmath27 of the incoming light in the actual optical trap . however , this low efficiency does not limit the application of the optical traps . whilst we have used a limited power for diagnostics ,
our slm can be illuminated by up to @xmath28w@xmath29 .
we could , for example , illuminate the slm with @xmath30mw of @xmath24 nm light to generate a ring of radius @xmath31 m and waist @xmath32 m , giving us @xmath33mw in the measure region ; this corresponds to a ring trap of depth @xmath34nk for @xmath35rb atoms .
the light pattern shown here is two - dimensional , and disperses quickly away from the fourier plane . to confine atoms to the fourier plane we will add an orthogonal light sheet with a trapping frequency close to that of the ring trap .
the chemical potential of a bose
einstein condensate in a ring trap is calculated using the thomas - fermi approximation to be @xmath36 where @xmath9 is the radius of the ring as before , @xmath37hz is the radial trapping frequency , @xmath38 is the number of trapped atoms and @xmath39 is the s - wave scattering length . for @xmath40 @xmath35rb atoms in the ring trap described above
, the chemical potential is @xmath41nk .
this is around one third of the trap depth given above , thus making trapping possible .
whilst the computed patterns give smooth and accurate trapping potentials , imperfect device response and aberrations cause significant discrepancies and roughness in the resulting light pattern shown in figure [ fig : outputs]a .
we improve the light pattern s suitability as an atom trap by including the algorithm in a feedback loop .
placing a ccd camera ( thorlabs dcu200 series ) in the fourier plane of a @xmath42 mm lens allows us to measure the difference between the target profile and the real recorded intensity pattern .
we then create a new target pattern by subtracting the discrepancy from the original target , and re - run the mraf algorithm in an attempt to minimise this difference .
after a few iterations , the measured light profile shown in figure [ fig : outputs]b is noticeably closer to the original target . to illustrate this
, we plot the intensity of light around the trapping minimum of a section of the ring by integrating over the pixels within a small angle , as shown in figure [ fig : outputs]c .
it follows from equation [ eqn : trappingpot ] that any roughness in the intensity of the light pattern produces a roughness in the energy landscape , which may in turn cause heating or fragmentation of the atom cloud .
our feedback loop reduce the rms error from @xmath43 to @xmath44 at the minimum of our optical trap .
this gives us a roughness of @xmath45 , lower than the value reported in @xcite , which should allow us to observe superflow in this trap .
as experiments with cold atoms are carried out under vacuum , the windows of the vacuum system will themselves introduce aberrations which would not be accounted for by the feedback loop in its present form .
however , atoms are a very sensitive probe of roughness of the trapping potential , so we envisage continuing to use the feedback loop by taking absorption images of the trapped atoms rather than directly imaging the light profile , and use these images within the feedback loop .
we have shown that computer generated holographic techniques supplemented with a feedback algorithm can produce flexible , smooth , all - optical traps for the investigation of superfluidity and persistent currents in ultracold atoms .
the trap depth at different points of the pattern can be dynamically varied and the whole pattern can be rotated to induce superflow .
the geometry created using this method need not be restricted to the one presented here : the creation of ring lattices or ring potentials with areas of lowered intensity for the investigation of josephson tunnelling , or increasingly non - trivial geometries , should be possible .
for example , by creating two of these traps which intersect and rotate in opposite directions , one could investigate an area of crossing superfluids .
an appropriate choice of trapping light wavelength will allow mixtures of ultracold bosons and fermions and their corresponding superfluidities to be investigated . improved understanding of superfluidity in ultracold atoms may help to solve more difficult problems in condensed matter physics , such as the much - sought explanation of high-@xmath46 superconductors .
this work is supported by the uk epsrc .
gs acknowledges support from a scottish universities physics alliance ( supa ) advanced fellowship .
the authors would like to acknowledge useful discussions with vincent boyer , tiffany harte and sarah bromley .
10 rhodes dp , gherardi dm , livesey j , mcgloin d , melville h , freegarde t and dholakia k 2006 atom guiding along high order laguerre - gaussian light beams formed by spatial light modulation _ j. mod .
opt . _ * 53 * 547556 | we discuss the suitability of holographically generated optical potentials for the investigation of superfluidity in ultracold atoms . by using a spatial light modulator and
a feedback enabled algorithm we generate a smooth ring with variable bright regions that can be dynamically rotated to stir ultracold atoms and induce superflow .
we also comment on its future integration into a cold atoms experiment . |
we study degrees of freedom , or the `` effective number of parameters , '' in @xmath2-penalized linear regression problems .
in particular , for a response vector @xmath3 , predictor matrix @xmath4 and tuning parameter @xmath5 , we consider the lasso problem [ @xcite , @xcite ] @xmath6 the above notation emphasizes the fact that the solution @xmath7 may not be unique [ such nonuniqueness can occur if @xmath8 . throughout the paper , when a function @xmath9 may have a nonunique minimizer over its domain @xmath1
, we write @xmath10 to denote the set of minimizing @xmath11 values , that is , @xmath12 .
a fundamental result on the degrees of freedom of the lasso fit was shown by @xcite .
the authors show that if @xmath13 follows a normal distribution with spherical covariance , @xmath14 , and @xmath15 are considered fixed with @xmath16 , then @xmath17 where @xmath18 denotes the active set of the unique lasso solution at @xmath13 , and @xmath19 is its cardinality .
this is quite a well - known result , and is sometimes used to informally justify an application of the lasso procedure , as it says that number of parameters used by the lasso fit is simply equal to the ( average ) number of selected variables .
however , we note that the assumption @xmath16 implies that @xmath20 ; in other words , the degrees of freedom result ( [ eq : lassodffull ] ) does not cover the important `` high - dimensional '' case @xmath21 . in this case
, the lasso solution is not necessarily unique , which raises the questions : * can we still express degrees of freedom in terms of the active set of a lasso solution ? *
if so , which active set ( solution ) would we refer to ? in section [ sec
: lasso ] , we provide answers to these questions , by proving a stronger result when @xmath0 is a general predictor matrix . we show that the subspace spanned by the columns of @xmath0 in @xmath22 is almost surely unique , where `` almost surely '' means for almost every @xmath3 .
furthermore , the degrees of freedom of the lasso fit is simply the expected dimension of this column space .
we also consider the generalized lasso problem , @xmath23 where @xmath24 is a penalty matrix , and again the notation emphasizes the fact that @xmath7 need not be unique [ when @xmath8 .
this of course reduces to the usual lasso problem ( [ eq : lasso ] ) when @xmath25 , and @xcite demonstrate that the formulation ( [ eq : genlasso ] ) encapsulates several other important problems including the fused lasso on any graph and trend filtering of any order by varying the penalty matrix @xmath1 .
the same paper shows that if @xmath13 is normally distributed as above , and @xmath26 are fixed with @xmath16 , then the generalized lasso fit has degrees of freedom @xmath27.\ ] ] here @xmath28 denotes the boundary set of an optimal subgradient to the generalized lasso problem at @xmath13 ( equivalently , the boundary set of a dual solution at @xmath13 ) , @xmath29 denotes the matrix @xmath1 after having removed the rows that are indexed by @xmath30 , and @xmath31 , the dimension of the null space of @xmath29 .
it turns out that examining ( [ eq : genlassodffull ] ) for specific choices of @xmath1 produces a number of interpretable corollaries , as discussed in @xcite .
for example , this result implies that the degrees of freedom of the fused lasso fit is equal to the expected number of fused groups , and that the degrees of freedom of the trend filtering fit is equal to the expected number of knots @xmath32 @xmath33 , where @xmath34 is the order of the polynomial . the result ( [ eq : genlassodffull ] ) assumes that @xmath16 and does not cover the case @xmath21 ; in section [ sec : genlasso ] , we derive the degrees of freedom of the generalized lasso fit for a general @xmath0 ( and still a general @xmath1 ) . as in the lasso case
, we prove that there exists a linear subspace @xmath35 that is almost surely unique , meaning that it will be the same under different boundary sets @xmath30 corresponding to different solutions of ( [ eq : genlasso ] ) .
the generalized lasso degrees of freedom is then the expected dimension of this subspace .
our assumptions throughout the paper are minimal . as was already mentioned , we place no assumptions whatsoever on the predictor matrix @xmath36 or on the penalty matrix @xmath24 , considering them fixed and nonrandom .
we also consider @xmath37 fixed . for theorems [ thm : lassodfequi ] , [ thm : lassodfact ] and [ thm : genlassodf ]
we assume that @xmath13 is normally distributed , @xmath38 for some ( unknown ) mean vector @xmath39 and marginal variance @xmath40 .
this assumption is only needed in order to apply stein s formula for degrees of freedom , and none of the other lasso and generalized lasso results in the paper , namely lemmas [ lem : lassoproj ] through [ lem : invbound ] , make any assumption about the distribution of @xmath13 .
this paper is organized as follows .
the rest of the introduction contains an overview of related work , and an explanation of our notation .
section [ sec : prelim ] covers some relevant background material on degrees of freedom and convex polyhedra .
though the connection may not be immediately obvious , the geometry of polyhedra plays a large role in understanding problems ( [ eq : lasso ] ) and ( [ eq : genlasso ] ) , and section [ sec : poly ] gives a high - level view of this geometry before the technical arguments that follow in sections [ sec : lasso ] and [ sec : genlasso ] . in section [ sec : lasso ] , we derive two representations for the degrees of freedom of the lasso fit , given in theorems [ thm : lassodfequi ] and [ thm : lassodfact ] . in section [ sec : genlasso ] , we derive the analogous results for the generalized lasso problem , and these are given in theorem [ thm : genlassodf ] . as the lasso problem is a special case of the generalized lasso problem ( corresponding to @xmath25 ) , theorems [ thm : lassodfequi ] and [ thm : lassodfact ] can actually be viewed as corollaries of theorem [ thm : genlassodf ] .
the reader may then ask : why is there a separate section dedicated to the lasso problem ?
we give two reasons : first , the lasso arguments are simpler and easier to follow than their generalized lasso counterparts ; second , we cover some intermediate results for the lasso problem that are interesting in their own right and that do not carry over to the generalized lasso perspective .
section [ sec : disc ] contains some final discussion .
all of the degrees of freedom results discussed here assume that the response vector has distribution @xmath14 , and that the predictor matrix @xmath0 is fixed .
to the best of our knowledge , @xcite were the first to prove a result on the degrees of freedom of the lasso fit , using the lasso solution path with @xmath41 moving from @xmath42 to @xmath43 .
the authors showed that when the active set reaches size @xmath34 along this path , the lasso fit has degrees of freedom exactly @xmath34 .
this result assumes that @xmath0 has full column rank and further satisfies a restrictive condition called the `` positive cone condition , '' which ensures that as @xmath41 decreases , variables can only enter , and not leave , the active set .
subsequent results on the lasso degrees of freedom ( including those presented in this paper ) differ from this original result in that they derive degrees of freedom for a fixed value of the tuning parameter @xmath41 , and not a fixed number of steps @xmath34 taken along the solution path .
as mentioned previously , @xcite established the basic lasso degrees of freedom result ( for fixed @xmath41 ) stated in ( [ eq : lassodffull ] ) .
this is analogous to the path result of @xcite ; here degrees of freedom is equal to the expected size of the active set ( rather than simply the size ) because for a fixed @xmath41 the active set is a random quantity , and can hence achieve a random size .
the proof of ( [ eq : lassodffull ] ) appearing in @xcite relies heavily on properties of the lasso solution path . as also mentioned previously , @xcite derived an extension of ( [ eq : lassodffull ] ) to the generalized lasso problem , which is stated in ( [ eq : genlassodffull ] ) for an arbitrary penalty matrix @xmath1 .
their arguments are not based on properties of the solution path , but instead come from a geometric perspective much like the one developed in this paper .
both of the results ( [ eq : lassodffull ] ) and ( [ eq : genlassodffull ] ) assume that @xmath16 ; the current work extends these to the case of an arbitrary matrix @xmath0 , in theorems [ thm : lassodfequi ] , [ thm : lassodfact ] ( the lasso ) and [ thm : genlassodf ] ( the generalized lasso ) . in terms of our intermediate results ,
a version of lemmas [ lem : lcequi ] , [ lem : lcact ] corresponding to @xmath16 appears in @xcite , and a version of lemma [ lem : lcbound ] corresponding to @xmath16 appears in @xcite [ furthermore , @xcite only consider the boundary set representation and not the active set representation ] .
lemmas [ lem : nonexp ] , [ lem : locaff ] and the conclusions thereafter , on the degrees of freedom of the projection map onto a convex polyhedron , are essentially given in @xcite , though these authors state and prove the results in a different manner . in preparing a draft of this manuscript ,
it was brought to our attention that other authors have independently and concurrently worked to extend results ( [ eq : lassodffull ] ) and ( [ eq : genlassodffull ] ) to the general @xmath0 case .
namely , @xcite prove a result on the lasso degrees of freedom , and @xcite prove a result on the generalized lasso degrees of freedom , both for an arbitrary @xmath0 .
these authors results express degrees of freedom in terms of the active sets of special ( lasso or generalized lasso ) solutions .
theorems [ thm : lassodfact ] and [ thm : genlassodf ] express degrees of freedom in terms of the active sets of any solutions , and hence the appropriate application of these theorems provides an alternative verification of these formulas .
we discuss this in detail in the form of remarks following the theorems . in this paper ,
we use @xmath44 , @xmath45 and @xmath46 to denote the column space , row space and null space of a matrix @xmath47 , respectively ; we use @xmath48 and @xmath49 to denote the dimensions of @xmath44 [ equivalently , @xmath45 ] and @xmath46 , respectively .
we write @xmath51 for the the moore
penrose pseudoinverse of @xmath47 ; for a rectangular matrix @xmath47 , recall that @xmath52 .
we write @xmath53 to denote the projection matrix onto a linear subspace @xmath54 , and more generally , @xmath55 to denote the projection of a point @xmath11 onto a closed convex set @xmath56 . for readability ,
we sometimes write @xmath57 ( instead of @xmath58 ) to denote the inner product between vectors @xmath59 and @xmath60 . for a set of indices
@xmath61 satisfying @xmath62 , and a vector @xmath63 , we use @xmath64 to denote the subvector @xmath65 . we denote the complementary subvector by @xmath66 .
the notation is similar for matrices . given another subset of indices @xmath67 with @xmath68 , and a matrix @xmath69 , we use @xmath70 to denote the submatrix @xmath71 \in{\mathbb{r}}^{k \times\ell}.\ ] ] in words , rows are indexed by @xmath72 , and columns are indexed by @xmath73 . when combining this notation with the transpose operation , we assume that the indexing happens first , so that @xmath74 . as above ,
negative signs are used to denote the complementary set of rows or columns ; for example , @xmath75 . to extract only rows or only columns , we abbreviate the other dimension by a dot , so that @xmath76 and @xmath77 ; to extract a single row or column , we use @xmath78 or @xmath79 . finally , and most importantly , we introduce the following shorthand notation : * for the predictor matrix @xmath80 , we let @xmath81 . * for the penalty matrix @xmath24 , we let @xmath82 . in other words ,
the default for @xmath0 is to index its columns , and the default for @xmath1 is to index its rows .
this convention greatly simplifies the notation in expressions that involve multiple instances of @xmath83 or @xmath84 ; however , its use could also cause a great deal of confusion , if not properly interpreted by the reader !
the following two sections describe some background material needed to follow the results in sections [ sec : lasso ] and [ sec : genlasso ] .
if the data vector @xmath3 is distributed according to the homoskedastic model @xmath85 , meaning that the components of @xmath13 are uncorrelated , with @xmath86 having mean @xmath87 and variance @xmath88 for @xmath89 , then the degrees of freedom of a function @xmath90 with @xmath91 , is defined as @xmath92 this definition is often attributed to @xcite or @xcite , and is interpreted as the `` effective number of parameters '' used by the fitting procedure @xmath93 .
note that for the linear regression fit of @xmath3 onto a fixed and full column rank predictor matrix @xmath36 , we have @xmath94 , and @xmath95 , which is the number of fitted coefficients ( one for each predictor variable )
. furthermore , we can decompose the risk of @xmath96 , denoted by @xmath97 , as @xmath98 a well - known identity that leads to the derivation of the @xmath99 statistic [ @xcite ] . for a general fitting procedure
@xmath93 , the motivation for the definition ( [ eq : df ] ) comes from the analogous decomposition of the quantity @xmath100 , @xmath101 therefore a large difference between risk and expected training error implies a large degrees of freedom .
why is the concept of degrees of freedom important ?
one simple answer is that it provides a way to put different fitting procedures on equal footing .
for example , it would not seem fair to compare a procedure that uses an effective number of parameters equal to 100 with another that uses only 10 .
however , assuming that these procedures can be tuned to varying levels of adaptivity ( as is the case with the lasso and generalized lasso , where the adaptivity is controlled by @xmath41 ) , one could first tune the procedures to have the same degrees of freedom , and then compare their performances .
doing this over several common values for degrees of freedom may reveal , in an informal sense , that one procedure is particularly efficient when it comes to its parameter usage versus another . a more detailed answer to the above question
is based the risk decomposition ( [ eq : riskd ] ) .
the decomposition suggests that an estimate @xmath102 of degrees of freedom can be used to form an estimate of the risk , @xmath103 furthermore , it is straightforward to check that an unbiased estimate of degrees of freedom leads to an unbiased estimate of risk ; that is , @xmath104 $ ] implies @xmath105 $ ] .
hence , the risk estimate ( [ eq : riskhat ] ) can be used to choose between fitting procedures , assuming that unbiased estimates of degrees of freedom are available .
[ it is worth mentioning that bootstrap or monte carlo methods can be helpful in estimating degrees of freedom ( [ eq : df ] ) when an analytic form is difficult to obtain . ]
the natural extension of this idea is to use the risk estimate ( [ eq : riskhat ] ) for tuning parameter selection . if we suppose that @xmath93 depends on a tuning parameter @xmath106 , denoted @xmath107 , then in principle one could minimize the estimated risk over @xmath41 to select an appropriate value for the tuning parameter , @xmath108 this is a computationally efficient alternative to selecting the tuning parameter by cross - validation , and it is commonly used ( along with similar methods that replace the factor of @xmath109 above with a function of @xmath110 or @xmath111 ) in penalized regression problems . even though such an estimate ( [ eq : tunsel ] ) is commonly used in the high - dimensional setting ( @xmath21 ) , its asymptotic properties are largely unknown in this case , such as risk consistency , or relatively efficiency compared to the cross - validation estimate .
@xcite proposed the risk estimate ( [ eq : riskhat ] ) using a particular unbiased estimate of degrees of freedom , now commonly referred to as _ stein s unbiased risk estimate _ ( sure ) .
stein s framework requires that we strengthen our distributional assumption on @xmath13 and assume normality , as stated in ( [ eq : normal ] ) .
we also assume that the function @xmath93 is continuous and almost differentiable .
( the precise definition of almost differentiability is not important here , but the interested reader may take it to mean that each coordinate function @xmath112 is absolutely continuous on almost every line segment parallel to one of the coordinate axes . ) given these assumptions , stein s main result is an alternate expression for degrees of freedom , @xmath113,\ ] ] where the function @xmath114 is called the divergence of @xmath93 .
immediately following is the unbiased estimate of degrees of freedom , @xmath115 we pause for a moment to reflect on the importance of this result . from its definition ( [ eq : df ] ) , we can see that the two most obvious candidates for unbiased estimates of degrees of freedom are @xmath116 \bigr ) y_i.\ ] ] to use the first estimate above , we need to know @xmath117 ( remember , this is ultimately what we are trying to estimate ! ) . using the second requires knowing @xmath118 $ ] , which is equally impractical because this invariably depends on @xmath117 .
on the other hand , stein s unbiased estimate ( [ eq : steindfhat ] ) does not have an explicit dependence on @xmath117 ; moreover , it can be analytically computed for many fitting procedures @xmath93 .
for example , theorem [ thm : lassodfact ] in section [ sec : lasso ] shows that , except for @xmath13 in a set of measure zero , the divergence of the lasso fit is equal to @xmath119 with @xmath18 being the active set of a lasso solution at @xmath13 .
hence , stein s formula allows for the unbiased estimation of degrees of freedom ( and subsequently , risk ) for a broad class of fitting procedures @xmath93something that may have not seemed possible when working from the definition directly .
a set @xmath120 is called a _ convex polyhedron _ , or simply a _
polyhedron _ , if @xmath56 is the intersection of finitely many half - spaces , @xmath121 where @xmath122 and @xmath123 .
( note that we do not require boundedness here ; a bounded polyhedron is sometimes called a polytope . )
see figure [ fig : poly ] for an example . there is a rich theory on polyhedra ; the definitive reference is @xcite , and another good reference is @xcite . as this is a paper on statistics and not geometry , we do not attempt to give an extensive treatment of the properties of polyhedra .
we do , however , give two properties ( in the form of two lemmas ) that are especially important with respect to our statistical problem ; our discussion will also make it clear why polyhedra are relevant in the first place . .
_ _ ] from its definition ( [ eq : poly ] ) , it follows that a polyhedron is a closed convex set . the first property that we discuss does
not actually rely on the special structure of polyhedra , but only on convexity .
for any closed convex set
@xmath120 and any point @xmath124 , there is a unique point @xmath125 minimizing @xmath126 . to see this , note that if @xmath127 is another minimizer , @xmath128 , then by convexity @xmath129 , and @xmath130 , a contradiction .
therefore , the projection map onto @xmath56 is indeed well defined , and we write this as @xmath131 , @xmath132 for the usual linear regression problem , where @xmath3 is regressed onto @xmath36 , the fit @xmath133 can be written in terms of the projection map onto the polyhedron @xmath134 , as in @xmath135 . furthermore , for both the lasso and generalized lasso problems , ( [ eq : lasso ] ) and ( [ eq : genlasso ] ) , it turns out that we can express the fit as the residual from projecting onto a suitable polyhedron @xmath136 , that is , @xmath137 this is proved in lemma [ lem : lassoproj ] for the lasso and in lemma [ lem : genlassoproj ] for the generalized lasso ( the polyhedron @xmath56 depends on @xmath15 for the lasso case , and on @xmath26 for the generalized lasso case ) .
our first lemma establishes that both the projection map onto a closed convex set and the residual map are nonexpansive , hence continuous and almost differentiable everywhere .
these are the conditions needed to apply stein s formula .
[ lem : nonexp ] for any closed convex set @xmath120 , both the projection map @xmath138 and the residual projection map @xmath139 are nonexpansive .
that is , they satisfy @xmath140 therefore , @xmath141 and @xmath142 are both continuous and almost differentiable .
the proof can be found in appendix [ app : nonexp ] .
lemma [ lem : nonexp ] will be quite useful later in the paper , as it will allow us to use stein s formula to compute the degrees of freedom of the lasso and generalized lasso fits , after showing that these fits are indeed the residuals from projecting onto closed convex sets . the second property that we discuss uses the structure of polyhedra .
unlike lemma [ lem : nonexp ] , this property will not be used directly in the following sections of the paper ; instead , we present it here to give some intuition with respect to the degrees of freedom calculations to come .
the property can be best explained by looking back at figure [ fig : poly ] .
loosely speaking , the picture suggests that we can move the point @xmath11 around a bit and it will still project to the same face of @xmath56 .
another way of saying this is that there is a neighborhood of @xmath11 on which @xmath141 is simply the projection onto an affine subspace
. this would not be true if @xmath11 is in some exceptional set , which is made up of rays that emanate from the corners of @xmath56 , like the two drawn in the bottom right corner of figure .
however , the union of such rays has measure zero , so the map @xmath141 is locally an affine projection , almost everywhere .
this idea can be stated formally as follows .
[ lem : locaff ] let @xmath120 be a polyhedron . for almost every @xmath143 , there is an associated neighborhood @xmath144 of @xmath11 , linear subspace @xmath145 and point @xmath146 , such that the projection map restricted to @xmath144 , @xmath147 , is @xmath148 which is simply the projection onto the affine subspace @xmath149 . the proof is given in appendix [ app : locaff ] . these last two properties can be used to derive a general expression for the degrees of freedom of the fitting procedure @xmath150 , when @xmath120 is a polyhedron .
[ a similar formula holds for @xmath151 .
] lemma [ lem : nonexp ] tells us that @xmath142 is continuous and almost differentiable , so we can use stein s formula ( [ eq : steindf ] ) to compute its degrees of freedom .
lemma [ lem : locaff ] tells us that for almost every @xmath152 , there is a neighborhood @xmath144 of @xmath13 , linear subspace @xmath153 , and point @xmath146 , such that @xmath154 therefore , @xmath155 and an expectation over @xmath13 gives @xmath156.\ ] ] it should be made clear that the random quantity in the above expectation is the linear subspace @xmath157 , which depends on @xmath13 . in a sense ,
the remainder of this paper is focused on describing @xmath158the dimension of the face of @xmath56 onto which the point @xmath13 projects in a meaningful way for the lasso and generalized lasso problems .
section [ sec : lasso ] considers the lasso problem , and we show that @xmath54 can be written in terms of the equicorrelation set of the fit at @xmath13 . we also show that @xmath54 can be described in terms of the active set of a solution at @xmath13 . in section [ sec : genlasso ] we show the analogous results for the generalized lasso problem , namely , that @xmath54 can be written in terms of either the boundary set of an optimal subgradient at @xmath13 ( the analogy of the equicorrelation set for the lasso ) or the active set of a solution at @xmath13 .
in this section we derive the degrees of freedom of the lasso fit , for a general predictor matrix @xmath0 .
all of our arguments stem from the karush
tucker ( kkt ) optimality conditions , and we present these first .
we note that many of the results in this section can be alternatively derived using the lasso dual problem .
appendix [ app : dual ] explains this connection more precisely .
for the current work , we avoid the dual perspective simply to keep the presentation more self - contained .
finally , we remind the reader that @xmath83 is used to extract columns of @xmath0 corresponding to an index set @xmath73.=1 the kkt conditions for the lasso problem ( [ eq : lasso ] ) can be expressed as @xmath159 & \quad if $ { { \hat{\beta}}}_i = 0$. } & \ ] ] here @xmath160 is a subgradient of the function @xmath161 evaluated at @xmath162 .
hence @xmath7 is a minimizer in ( [ eq : lasso ] ) if and only if @xmath7 satisfies ( [ eq : lassokkt ] ) and ( [ eq : lassosg ] ) for some @xmath163 . directly from the kkt conditions , we can show that @xmath133 is the residual from projecting @xmath13 onto a polyhedron .
[ lem : lassoproj ] for any @xmath0 and @xmath5 , the lasso fit @xmath133 can be written as @xmath164 , where @xmath165 is the polyhedron @xmath166 given a point @xmath3 , its projection @xmath167 onto a closed convex set @xmath120 can be characterized as the unique point satisfying @xmath168 hence defining @xmath169 , and @xmath56 as in the lemma , we want to show that ( [ eq : opt ] ) holds for all @xmath125 .
well , @xmath170 \\[-8pt ] & = & \langle x{{\hat{\beta } } } , y - x{{\hat{\beta}}}\rangle - \langle x{^t}u , { { \hat{\beta}}}\rangle.\nonumber\end{aligned}\ ] ] consider the first term above . taking an inner product with @xmath7 on both sides of ( [ eq : lassokkt ] )
gives @xmath171 .
furthermore , the @xmath2 norm can be characterized in terms of its dual norm , the @xmath172 norm , as in @xmath173 therefore , continuing from ( [ eq : innerprod ] ) , we have @xmath174 which is @xmath175 for all @xmath125 , and we have hence proved that @xmath176 . to show that @xmath56 is indeed a polyhedron , note that it can be written as @xmath177 which is a finite intersection of half - spaces
. showing that the lasso fit is the residual from projecting @xmath13 onto a polyhedron is important , because it means that @xmath178 is nonexpansive as a function of @xmath13 , and hence continuous and almost differentiable , by lemma [ lem : nonexp ] .
this establishes the conditions that are needed to apply stein s formula for degrees of freedom . in the next section ,
we define the equicorrelation set @xmath179 , and show that the lasso fit and solutions both have an explicit form in terms of @xmath179 . following this
, we derive an expression for the lasso degrees of freedom as a function of the equicorrelation set . according to lemma [ lem : lassoproj ] ,
the lasso fit @xmath133 is always unique ( because projection onto a closed convex set is unique ) .
therefore , even though the solution @xmath7 is not necessarily unique , the optimal subgradient @xmath163 is unique , because it can be written entirely in terms of @xmath133 , as shown by ( [ eq : lassokkt ] ) . we define the unique _ equicorrelation set _
@xmath179 as @xmath180 an alternative definition for the equicorrelation set is @xmath181 which explains its name , as @xmath179 can be thought of as the set of variables that have equal and maximal absolute inner product ( or correlation for standardized variables ) with the residual .
the set @xmath179 is a natural quantity to work with because we can express the lasso fit and the set of lasso solutions in terms of @xmath179 , by working directly from equation ( [ eq : lassokkt ] ) .
first we let @xmath182 the signs of the inner products of the equicorrelation variables with the residual .
since @xmath183 by definition of the subgradient , the @xmath179 block of the kkt conditions can be rewritten as @xmath184 because @xmath185 , we can write @xmath186 , so rearranging ( [ eq : lassokkt2 ] ) we get @xmath187 therefore , the lasso fit @xmath188 is @xmath189 and any lasso solution must be of the form @xmath190 where @xmath191 . in the case
that @xmath192for example , this holds if @xmath16the lasso solution is unique and is given by ( [ eq : lassosol ] ) with @xmath193 .
but in general , when @xmath194 , it is important to note that not every @xmath191 necessarily leads to a lasso solution in ( [ eq : lassosol ] ) ; the vector @xmath60 must also preserve the signs of the nonzero coefficients ; that is , it must also satisfy @xmath195_i + b_i
\bigr ) = s_i\nonumber \\[-8pt ] \\[-8pt ] \eqntext{\mbox{for each $ i$ such that } \bigl[(x_{\mathcal{e}})^+ \bigl(y - ( x_{\mathcal{e}}{^t})^+ \lambda s \bigr ) \bigr]_i + b_i \not= 0.}\end{aligned}\ ] ] otherwise , @xmath163 would not be a proper subgradient of @xmath196 .
using relatively simple arguments , we can derive a result on the lasso degrees of freedom in terms of the equicorrelation set .
our arguments build on the following key lemma .
[ lem : lassosol0 ] for any @xmath197 and @xmath5 , a lasso solution is given by @xmath198 where @xmath179 and @xmath199 are the equicorrelation set and signs , as defined in ( [ eq : equiset ] ) and ( [ eq : equisigns ] ) .
in other words , lemma [ lem : lassosol0 ] says that the sign condition ( [ eq : lassosign ] ) is always satisfied by taking @xmath193 , regardless of the rank of @xmath0 .
this result is inspired by the lars work of @xcite , though it is not proved in the lars paper ; see appendix b of @xcite for a proof . next
we show that , almost everywhere in @xmath13 , the equicorrelation set and signs are locally constant functions of @xmath13 .
to emphasize their functional dependence on @xmath13 , we write them as @xmath200 and @xmath201 . [ lem : lcequi ] for almost every @xmath3 , there exists a neighborhood @xmath144 of @xmath13 such that @xmath202 and @xmath203 for all @xmath204 .
define @xmath205_{(i,\cdot ) } \bigl(z - ( x_{\mathcal{e}}{^t})^+ \lambda s \bigr ) = 0 \bigr\},\ ] ] where the first union above is taken over all subsets @xmath206 and sign vectors @xmath207 , but we exclude sets @xmath179 for which a row of @xmath208 is entirely zero .
the set @xmath209 is a finite union of affine subspaces of dimension @xmath210 , and therefore has measure zero .
let @xmath211 , and abbreviate the equicorrelation set and signs as @xmath212 and @xmath213
. we may assume no row of @xmath208 is entirely zero .
( otherwise , this implies that @xmath214 has a zero column , which implies that @xmath215 , a trivial case for this lemma . ) therefore , as @xmath211 , this means that the lasso solution given in ( [ eq : lassosol0 ] ) satisfies @xmath216 for every @xmath217 .
now , for a new point @xmath218 , consider defining @xmath219 we need to verify that this is indeed a solution at @xmath218 , and that the corresponding fit has equicorrelation set @xmath179 and signs @xmath199 .
first notice that , after a straightforward calculation , @xmath220 also , by the continuity of the function @xmath221 , @xmath222 there exists a neighborhood @xmath223 of @xmath13 such that @xmath224 for all @xmath225 .
hence @xmath226 has equicorrelation set @xmath227 and signs . to check that @xmath228 is a lasso solution at @xmath218
, we consider the function @xmath229 , @xmath230 the continuity of @xmath93 implies that there exists a neighborhood @xmath231 of @xmath13 such that @xmath232_i \not= 0 \qquad\mbox{for } i \in{\mathcal{e } } , \quad\mbox{and } \\ { \operatorname{sign}}({{\hat{\beta}}}_{\mathcal{e}}(y ' ) ) & = & { \operatorname{sign}}\bigl((x_{\mathcal{e}})^+ \bigl(y ' - ( x_{\mathcal{e}}{^t})^+ \lambda s \bigr ) \bigr)\end{aligned}\ ] ] for each @xmath233 .
defining @xmath234 completes the proof .
this immediately implies the following theorem .
[ thm : lassodfequi ] assume that @xmath13 follows a normal distribution ( [ eq : normal ] ) . for any @xmath0 and @xmath5 , the lasso fit @xmath133 has degrees of freedom @xmath235,\ ] ] where @xmath212 is the equicorrelation set of the lasso fit at @xmath13 . by lemmas [ lem : nonexp ] and [ lem : lassoproj ] we know that @xmath178 is continuous and almost differentiable , so we can use stein s formula ( [ eq : steindf ] ) for degrees of freedom .
by lemma [ lem : lcequi ] , we know that @xmath212 and @xmath213 are locally constant for all @xmath211 . therefore , taking the divergence of the fit in ( [ eq : lassofit ] ) , we get @xmath236 taking an expectation over @xmath13 ( and recalling that @xmath209 has measure zero ) gives the result . next
, we shift our focus to a different subset of variables : the active set @xmath22 .
unlike the equicorrelation set , the active set is not unique , as it depends on a particular choice of lasso solution .
though it may seem that such nonuniqueness could present complications , it turns out that all of the active sets share a special property ; namely , the linear subspace @xmath237 is the same for any choice of active set @xmath22 , almost everywhere in @xmath13 .
this invariance allows us to express the degrees of freedom of lasso fit in terms of the active set ( or , more precisely , any active set ) .
given a particular solution @xmath7 , we define the _ active set _ @xmath22 as @xmath238 this is also called the support of @xmath7 and written @xmath239 . from ( [ eq : lassosol ] ) , we can see that we always have @xmath240 , and different active sets @xmath22 can be formed by choosing @xmath191 to satisfy the sign condition ( [ eq : lassosign ] ) and also @xmath241_i + b_i = 0 \qquad\mbox{for } i \notin{{\mathcal{a}}}.\ ] ] if @xmath16 , then @xmath193 , so there is a unique active set , and furthermore @xmath242 for almost every @xmath3 ( in particular , this last statement holds for , where @xmath209 is the set of measure zero set defined in the proof of lemma [ lem : lcequi ] ) . for the signs of the coefficients of active variables , we write @xmath243 and we note that @xmath244 . by similar arguments as those used to derive expression ( [ eq : lassofit ] ) for the fit in section [ sec : lassoequi ] , the lasso fit can also be written as @xmath245 for the active set @xmath22 and signs @xmath246 of any lasso solution @xmath7 .
if we could take the divergence of the fit in the expression above , and simply ignore the dependence of @xmath22 and @xmath246 on @xmath13 ( treat them as constants ) , then this would give @xmath247 .
in the next section , we show that treating @xmath22 and @xmath246 as constants in ( [ eq : lassofit2 ] ) is indeed correct , for almost every @xmath13 .
this property then implies that the linear subspace @xmath237 is invariant under any choice of active set @xmath22 , almost everywhere in @xmath13 ; moreover , it implies that we can write the lasso degrees of freedom in terms of any active set .
we first establish a result on the local stability of @xmath248 and @xmath249 [ written in this way to emphasize their dependence on @xmath13 , through a solution @xmath250 .
[ lem : lcact ] there is a set @xmath251 , of measure zero , with the following property : for @xmath252 , and for any lasso solution @xmath253 with active set @xmath248 and signs @xmath249 , there is a neighborhood @xmath144 of @xmath13 such that every point @xmath204 yields a lasso solution @xmath228 with the same active set @xmath254 and the same active signs @xmath255 .
the proof is similar to that of lemma [ lem : lcequi ] , except that it is longer and somewhat more complicated , so it is delayed until appendix [ app : lcact ] .
combined with expression ( [ eq : lassofit2 ] ) for the lasso fit , lemma [ lem : lcact ] now implies an invariance of the subspace spanned by the active variables .
[ lem : invact ] for the same set @xmath251 as in lemma [ lem : lcact ] , and for any @xmath252 , the linear subspace @xmath237 is invariant under all sets @xmath18 defined in terms of a lasso solution @xmath253 at @xmath13 .
let @xmath252 , and let @xmath253 be a solution with active set @xmath18 and signs @xmath256 .
let @xmath144 be the neighborhood of @xmath13 as constructed in the proof of lemma [ lem : lcact ] ; on this neighborhood , solutions exist with active set @xmath22 and signs @xmath246 .
hence , recalling ( [ eq : lassofit2 ] ) , we know that for every @xmath257 , @xmath258 now suppose that @xmath259 and @xmath260 are the active set and signs of another lasso solution at @xmath13
. then , by the same arguments , there is a neighborhood @xmath261 of @xmath13 such that @xmath262 for all @xmath263 . by the uniqueness of the fit
, we have that for each @xmath264 , @xmath265 since @xmath266 is open , for any @xmath267 , there is an @xmath268 such that @xmath269 . plugging @xmath270 into the above equation
implies that @xmath271 , so @xmath272 .
a similar argument gives @xmath273 , completing the proof .
again , this immediately leads to the following theorem .
[ thm : lassodfact ] assume that @xmath13 follows a normal distribution ( [ eq : normal ] ) . for any @xmath0 and @xmath5 , the lasso fit @xmath133 has degrees of freedom @xmath274,\ ] ] where @xmath18 is the active set corresponding to any lasso solution @xmath253 at @xmath13 .
note : by lemma [ lem : invact ] , @xmath119 is an invariant quantity , not depending on the choice of active set ( coming from a lasso solution ) , for almost every @xmath13 .
this makes the above result well defined .
proof of theorem [ thm : lassodfact ] we can apply stein s formula ( [ eq : steindf ] ) for degrees of freedom , because @xmath178 is continuous and almost differentiable by lemmas [ lem : nonexp ] and [ lem : lassoproj ] . let @xmath18 and @xmath256 be the active set and active signs of a lasso solution at @xmath252 , with @xmath275 as in lemma [ lem : invact ] . by this same lemma , there exists a lasso solution with active set @xmath22 and signs @xmath246 at every point @xmath218 in some neighborhood @xmath144 of @xmath13 , and therefore , taking the divergence of the fit ( [ eq : lassofit2 ] ) , we get @xmath276 taking an expectation over @xmath13 completes the proof . the proof of lemma [ lem : lcact ] showed that , for almost every @xmath13 , the equicorrelation set @xmath179 is actually the active set @xmath22 of the particular lasso solution defined in ( [ eq : lassosol0 ] ) .
hence theorem [ thm : lassodfequi ] can be viewed as a corollary of theorem [ thm : lassodfact ] . when @xmath16 , the lasso solution is unique , and there is only one active set @xmath22 . and as the columns of @xmath0 are linearly independent , we have @xmath277 , so the result of theorem [ thm : lassodfact ] reduces to @xmath278 as shown in @xcite .
an interesting result on the lasso degrees of freedom was recently and independently obtained by @xcite .
their result states that , for a general @xmath0 , @xmath279 where @xmath280 is the smallest cardinality among all active sets of lasso solutions .
this actually follows from theorem [ thm : lassodfact ] , by noting that for any @xmath13 there exists a lasso solution whose active set @xmath259 corresponds to linear independent predictors @xmath281 , so @xmath282 [ e.g. , see theorem 3 in appendix b of @xcite ] , and furthermore , for almost every @xmath13 no active set can have a cardinality smaller than @xmath280 , as this would contradict lemma [ lem : invact ] .
consider the elastic net problem [ @xcite ] , @xmath283 where we now have two tuning parameters @xmath284 .
note that our notation above emphasizes the fact that there is always a unique solution to the elastic net criterion , regardless of the rank of @xmath0 .
this property ( among others , such as stability and predictive ability ) is considered an advantage of the elastic net over the lasso .
we can rewrite the elastic net problem ( [ eq : enet ] ) as a ( full column rank ) lasso problem , @xmath285 \beta \right\|_{2}^2 + \lambda_1 \|\beta\|_{1},\ ] ] and hence it can be shown ( although we omit the details ) that the degrees of freedom of the elastic net fit is @xmath286,\ ] ] where @xmath18 is the active set of the elastic net solution at @xmath13 .
it is often more appropriate to include an ( unpenalized ) intercept coefficient in the lasso model , yielding the problem @xmath287 where @xmath288 is the vector of all @xmath289s . defining @xmath290
, we note that the fit of problem ( [ eq : lassoint ] ) can be written as @xmath291 , and that @xmath7 solves the usual lasso problem @xmath292 now it follows ( again we omit the details ) that the fit of the lasso problem with intercept ( [ eq : lassoint ] ) has degrees of freedom @xmath293,\ ] ] where @xmath18 is the active set of a solution @xmath253 at @xmath13 ( these are the nonintercept coefficients ) .
in other words , the degrees of freedom is one plus the expected dimension of the subspace spanned by the active variables , once we have centered these variables .
a similar result holds for an arbitrary set of unpenalized coefficients , by replacing @xmath294 above with the projection onto the orthogonal complement of the column space of the unpenalized variables , and @xmath289 above with the dimension of the column space of the unpenalized variables . as mentioned in the , a nice feature of the full column rank result ( [ eq : lassodffull ] )
is its interpretability and its explicit nature .
the general result is also explicit in the sense that an unbiased estimate of degrees of freedom can be achieved by computing the rank of a given matrix . in terms of interpretability , when @xmath16 , the degrees of freedom of the lasso fit is @xmath295this says that , on average , the lasso `` spends '' the same number of parameters as does linear regression on @xmath19 linearly independent predictor variables .
fortunately , a similar interpretation is possible in the general case : we showed in theorem [ thm : lassodfact ] that for a general predictor matrix @xmath0 , the degrees of freedom of the lasso fit is @xmath296 $ ] , the expected dimension of the linear subspace spanned by the active variables .
meanwhile , for the linear regression problem @xmath297 where we consider @xmath22 fixed , the degrees of freedom of the fit is @xmath298 . in other words ,
the lasso adaptively selects a subset @xmath22 of the variables to use for a linear model of @xmath13 , but on average it only `` spends '' the same number of parameters as would linear regression on the variables in @xmath22 , if @xmath22 was pre - specified .
how is this possible ? broadly speaking
, the answer lies in the shrinkage due to the @xmath2 penalty . although the active set is chosen adaptively , the lasso does not estimate the active coefficients as aggressively as does the corresponding linear regression problem ( [ eq : lsa ] ) ; instead , they are shrunken toward zero , and this adjusts for the adaptive selection .
differing views have been presented in the literature with respect to this feature of lasso shrinkage . on the one hand , for example , @xcite point out that lasso estimates suffer from bias due to the shrinkage of large coefficients , and motivate the nonconvex _ scad _ penalty as an attempt to overcome this bias . on the other hand , for example
, loubes and massart ( @xcite ) discuss the merits of such shrunken estimates in model selection criteria , such as ( [ eq : tunsel ] ) . in the current context ,
the shrinkage due to the @xmath2 penalty is helpful in that it provides control over degrees of freedom . a more precise study of this idea is the topic of future work .
in this section we extend our degrees of freedom results to the generalized lasso problem , with an arbitrary predictor matrix @xmath0 and penalty matrix @xmath1 . as before ,
the kkt conditions play a central role , and we present these first . also , many results that follow have equivalent derivations from the perspective of the generalized lasso dual problem ; see appendix [ app : dual ] .
we remind the reader that @xmath84 is used to extract to extract rows of @xmath1 corresponding to an index set @xmath72 . the kkt conditions for the generalized lasso problem ( [ eq : genlasso ] ) are @xmath299 & \quad if $ ( d{{\hat{\beta}}})_i = 0$. } & \ ] ] now @xmath300 is a subgradient of the function @xmath301 evaluated at @xmath302 .
similar to what we showed for the lasso , it follows from the kkt conditions that the generalized lasso fit is the residual from projecting @xmath13 onto a polyhedron .
[ lem : genlassoproj ] for any @xmath0 and @xmath5 , the generalized lasso fit can be written as @xmath164 , where @xmath165 is the polyhedron @xmath303 the proof is quite similar to that of lemma [ lem : lassoproj ] .
as in ( [ eq : innerprod ] ) , we want to show that @xmath304 for all @xmath125 , where @xmath56 is as in the lemma . for the first term above
, we can take an inner product with @xmath7 on both sides of ( [ eq : genlassokkt ] ) to get @xmath305 , and furthermore , @xmath306 therefore ( [ eq : innerprod2 ] ) holds if @xmath307 for some @xmath308 , in other words , if @xmath309 . to show that @xmath56 is a polyhedron , note that we can write it as @xmath310 where @xmath311 is taken to mean the inverse image under the linear map @xmath312 , and @xmath313 , a hypercube in @xmath314 .
clearly @xmath315 is a polyhedron , and the image or inverse image of a polyhedron under a linear map is still a polyhedron . as with the lasso
, this lemma implies that the generalized lasso fit @xmath178 is nonexpansive , and therefore continuous and almost differentiable as a function of @xmath13 , by lemma [ lem : nonexp ] .
this is important because it allows us to use stein s formula when computing degrees of freedom . in the next section
we define the boundary set @xmath30 , and derive expressions for the generalized lasso fit and solutions in terms of @xmath30 .
the following section defines the active set @xmath22 in the generalized lasso context , and again gives expressions for the fit and solutions in terms of @xmath22 .
though neither @xmath30 nor @xmath22 are necessarily unique for the generalized lasso problem , any choice of @xmath30 or @xmath22 generates a special invariant subspace ( similar to the case for the active sets in the lasso problem ) .
we are subsequently able to express the degrees of freedom of the generalized lasso fit in terms of any boundary set @xmath30 , or any active set @xmath22 . like the lasso ,
the generalized lasso fit @xmath133 is always unique ( following from lemma [ lem : genlassoproj ] , and the fact that projection onto a closed convex set is unique ) .
however , unlike the lasso , the optimal subgradient @xmath163 in the generalized lasso problem is not necessarily unique . in particular , if @xmath316 , then the optimal subgradient @xmath163 is not uniquely determined by conditions ( [ eq : genlassokkt ] ) and ( [ eq : genlassosg ] ) . given a subgradient @xmath163 satisfying ( [ eq : genlassokkt ] ) and ( [ eq : genlassosg ] ) for some @xmath7 , we define the _ boundary set _ @xmath30 as @xmath317 this generalizes the notion of the equicorrelation set @xmath179 in the lasso problem [ though , as just noted , the set @xmath30 is not necessarily unique unless ] .
we also define @xmath318 now we focus on writing the generalized lasso fit and solutions in terms of @xmath30 and @xmath199 . abbreviating @xmath319 , note that we can expand @xmath320 .
therefore , multiplying both sides of ( [ eq : genlassokkt ] ) by @xmath321 yields @xmath322 since @xmath323 , we can write @xmath324 . also , we have @xmath325 by definition of @xmath30 , so @xmath326 .
these two facts allow us to rewrite ( [ eq : genlassokktb ] ) as @xmath327 and hence the fit @xmath328 is @xmath329 where we have un - abbreviated @xmath319 .
further , any generalized lasso solution is of the form @xmath330 where @xmath331 .
multiplying the above equation by @xmath29 , and recalling that @xmath325 , reveals that @xmath332 ; hence @xmath333 . in the case that @xmath334 , the generalized lasso solution is unique and is given by ( [ eq : genlassosol ] ) with @xmath193 .
this occurs when @xmath16 , for example .
otherwise , any @xmath335 gives a generalized lasso solution in ( [ eq : genlassosol ] ) as long as it also satisfies the sign condition @xmath336 necessary to ensure that @xmath163 is a proper subgradient of @xmath337 .
we define the _ active set _ of a particular solution @xmath7 as @xmath338 which can be alternatively expressed as @xmath339 .
if @xmath7 corresponds to a subgradient with boundary set @xmath30 and signs @xmath199 , then @xmath340 ; in particular , given @xmath30 and @xmath199 , different active sets @xmath22 can be generated by taking @xmath341 such that ( [ eq : genlassosign ] ) is satisfied , and also @xmath342 if @xmath16 , then @xmath193 , and there is only one active set @xmath22 ; however , in this case , @xmath22 can still be a strict subset of @xmath30 .
this is quite different from the lasso problem , wherein @xmath242 for almost every @xmath13 whenever @xmath16 .
[ note that in the generalized lasso problem , @xmath16 implies that @xmath22 is unique but implies nothing about the uniqueness of @xmath30this is determined by the rank of @xmath1 .
the boundary set @xmath30 is not necessarily unique if @xmath316 , and in this case we may have @xmath343 for some @xmath344 , which certainly implies that @xmath345 for any @xmath3 .
hence some boundary sets may not correspond to active sets at any @xmath13 .
] we denote the signs of the active entries in @xmath346 by @xmath347 and we note that @xmath244 .
following the same arguments as those leading up to the expression for the fit ( [ eq : genlassofit ] ) in section [ sec : genlassobound ] , we can alternatively express the generalized lasso fit as @xmath348 where @xmath22 and @xmath246 are the active set and signs of any solution .
computing the divergence of the fit in ( [ eq : genlassofit2 ] ) , and pretending that @xmath22 and @xmath246 are constants ( not depending on @xmath13 ) , gives @xmath349 . the same logic applied to ( [ eq : genlassofit ] ) gives @xmath350 .
the next section shows that , for almost every @xmath13 , the quantities @xmath351 or @xmath352 can indeed be treated as locally constant in expressions ( [ eq : genlassofit2 ] ) or ( [ eq : genlassofit ] ) , respectively .
we then prove that linear subspaces @xmath353 are invariant under all choices of boundary sets @xmath30 , respectively active sets @xmath22 , and that the two subspaces are in fact equal , for almost every @xmath13 .
furthermore , we express the generalized lasso degrees of freedom in terms of any boundary set or any active set .
we call @xmath354 an _ optimal pair _ provided that @xmath355 and @xmath253 jointly satisfy the kkt conditions , ( [ eq : genlassokkt ] ) and ( [ eq : genlassosg ] ) , at @xmath13 .
for such a pair , we consider its boundary set @xmath356 , boundary signs @xmath201 , active set @xmath248 , active signs @xmath249 , and show that these sets and sign vectors possess a kind of local stability .
[ lem : lcbound ] there exists a set @xmath357 , of measure zero , with the following property : for @xmath211 , and for any optimal pair @xmath354 with boundary set @xmath356 , boundary signs @xmath201 , active set @xmath248 , and active signs @xmath249 , there is a neighborhood @xmath144 of @xmath13 such that each point @xmath257 yields an optimal pair @xmath358 with the same boundary set @xmath359 , boundary signs @xmath360 , active set @xmath254 and active signs @xmath255 .
the proof is delayed to appendix [ app : lcbound ] , mainly because of its length .
now lemma [ lem : lcbound ] , used together with expressions ( [ eq : genlassofit ] ) and ( [ eq : genlassofit2 ] ) for the generalized lasso fit , implies an invariance in representing a ( particularly important ) linear subspace .
[ lem : invbound ] for the same set @xmath357 as in lemma [ lem : lcbound ] , and for any @xmath211 , the linear subspace @xmath361 is invariant under all boundary sets @xmath28 defined in terms of an optimal subgradient at @xmath355 at @xmath13 .
the linear subspace @xmath362 is also invariant under all choices of active sets @xmath18 defined in terms of a generalized lasso solution @xmath253 at @xmath13 .
finally , the two subspaces are equal , @xmath363 .
let @xmath211 , and let @xmath355 be an optimal subgradient with boundary set @xmath28 and signs @xmath213 .
let @xmath144 be the neighborhood of @xmath13 over which optimal subgradients exist with boundary set @xmath30 and signs @xmath199 , as given by lemma [ lem : lcbound ] .
recalling the expression for the fit ( [ eq : genlassofit ] ) , we have that for every @xmath204 @xmath364 if @xmath253 is a solution with active set @xmath18 and signs @xmath256 , then again by lemma [ lem : lcbound ] there is a neighborhood @xmath365 of @xmath13 such that each point @xmath366 yields a solution with active set @xmath22 and signs @xmath246 .
[ note that @xmath365 and @xmath144 are not necessarily equal unless @xmath355 and @xmath253 jointly satisfy the kkt conditions at @xmath13 . ] therefore , recalling ( [ eq : genlassofit ] ) , we have @xmath367 for each @xmath366 .
the uniqueness of the generalized lasso fit now implies that @xmath368 for all @xmath369 .
as @xmath370 is open , for any @xmath371 , there exists an @xmath268 such that @xmath372 . plugging @xmath270 into the equation above
reveals that @xmath373 , hence @xmath374 .
the reverse inclusion follows similarly , and therefore@xmath375 .
finally , the same strategy can be used to show that these linear subspaces are unchanged for any choice of boundary set @xmath28 , coming from an optimal subgradient at @xmath13 and for any choice of active set @xmath18 coming from a solution at @xmath13 . noticing that @xmath376 for matrices @xmath377 gives the result as stated in the lemma .
this local stability result implies the following theorem .
[ thm : genlassodf ] assume that @xmath13 follows a normal distribution ( [ eq : normal ] ) . for any @xmath378 and @xmath5
, the degrees of freedom of the generalized lasso fit can be expressed as @xmath379,\ ] ] where @xmath28 is the boundary set corresponding to any optimal subgradient @xmath355 of the generalized lasso problem at @xmath13 .
we can alternatively express degrees of freedom as @xmath380,\ ] ] with @xmath18 being the active set corresponding to any generalized lasso solution @xmath253 at @xmath13 .
note : lemma [ lem : invbound ] implies that for almost every @xmath152 , for any @xmath30 defined in terms of an optimal subgradient , and for any @xmath22 defined in terms of a generalized lasso solution , @xmath381 .
this makes the above expressions for degrees of freedom well defined .
proof of theorem [ thm : genlassodf ] first , the continuity and almost differentiability of @xmath178 follow from lemmas [ lem : nonexp ] and [ lem : genlassoproj ] , so we can use stein s formula ( [ eq : steindf ] ) for degrees of freedom .
let @xmath211 , where @xmath209 is the set of measure zero as in lemma [ lem : lcact ] .
if @xmath28 and @xmath213 are the boundary set and signs of an optimal subgradient at @xmath13 , then by lemma [ lem : invbound ] there is a neighborhood @xmath144 of @xmath13 such that each point @xmath204 yields an optimal subgradient with boundary set @xmath30 and signs @xmath199 .
therefore , taking the divergence of the fit in ( [ eq : genlassofit ] ) , @xmath382 and taking an expectation over @xmath13 gives the first expression in the theorem .
similarly , if @xmath18 and @xmath256 are the active set and signs of a generalized lasso solution at @xmath13 , then by lemma [ lem : invbound ] there exists a solution with active set @xmath22 and signs @xmath246 at each point @xmath218 in some neighborhood @xmath365 of @xmath13 .
the divergence of the fit in ( [ eq : genlassofit2 ] ) is hence @xmath383 and taking an expectation over @xmath13 gives the second expression .
if @xmath16 , then @xmath384 for any linear subspace @xmath54 , so the results of theorem [ thm : genlassodf ] reduce to @xmath385 = { \mathrm{e}}[{\operatorname{nullity}}(d_{-{{\mathcal{a}}}})].\ ] ] the first equality above was shown in @xcite . analyzing the null space of @xmath29 ( equivalently , @xmath386 ) for specific choices of @xmath1
then gives interpretable results on the degrees of freedom of the fused lasso and trend filtering fits as mentioned in the introduction .
it is important to note that , as @xmath16 , the active set @xmath22 is unique , but not necessarily equal to the boundary set @xmath30 [ since @xmath30 can be nonunique if @xmath316 ] . if @xmath25 , then @xmath388 for any subset @xmath389 .
therefore the results of theorem [ thm : genlassodf ] become @xmath390 = { \mathrm{e}}[{\operatorname{rank}}(x_{{\mathcal{a}}})],\ ] ] which match the results of theorems [ thm : lassodfequi ] and [ thm : lassodfact ] ( recall that for the lasso the boundary set @xmath30 is exactly the same as equicorrelation set @xmath179 ) .
recent and independent work of @xcite shows that , for arbitrary @xmath378 and for any @xmath13 , there exists a generalized lasso solution whose active set @xmath259 satisfies @xmath391 ( calling @xmath259 the `` smallest '' active set is somewhat of an abuse of terminology , but it is the smallest in terms of the above intersection . ) the authors then prove that , for any @xmath378 , the generalized lasso fit has degrees of freedom @xmath392,\ ] ] with @xmath259 the special active set as above .
this matches the active set result of theorem [ thm : genlassodf ] applied to @xmath259 , since @xmath393 for this special active set .
we conclude this section by comparing the active set result of theorem [ thm : genlassodf ] to degrees of freedom in a particularly relevant equality constrained linear regression problem ( this comparison is similar to that made in lasso case , given at the end of section [ sec : lasso ] ) .
the result states that the generalized lasso fit has degrees of freedom @xmath394 $ ] , where @xmath18 is the active set of a generalized lasso solution at @xmath13 .
in other words , the complement of @xmath22 gives the rows of @xmath1 that are orthogonal to some generalized lasso solution .
now , consider the equality constrained linear regression problem @xmath395 in which the set @xmath22 is fixed .
it is straightforward to verify that the fit of this problem is the projection map onto @xmath396 , and hence has degrees of freedom @xmath397 .
this means that the generalized lasso fits a linear model of @xmath13 , and simultaneously makes the coefficients orthogonal to an adaptive subset @xmath22 of the rows of @xmath1 , but on average it only uses the same number of parameters as does the corresponding equality constrained linear regression problem ( [ eq : lsb ] ) , in which @xmath22 is pre - specified .
this seemingly paradoxical statement can be explained by the shrinkage due to the @xmath2 penalty .
even though the active set @xmath22 is chosen adaptively based on @xmath13 , the generalized lasso does not estimate the coefficients as aggressively as does the equality constrained linear regression problem ( [ eq : lsb ] ) , but rather , it shrinks them toward zero . roughly speaking
, his shrinkage can be viewed as a `` deficit '' in degrees of freedom , which makes up for the `` surplus '' attributed to the adaptive selection .
we study this idea more precisely in a future paper .
we showed that the degrees of freedom of the lasso fit , for an arbitrary predictor matrix @xmath0 , is equal to @xmath296 $ ] . here
@xmath18 is the active set of any lasso solution at @xmath13 , that is , @xmath398 .
this result is well defined , since we proved that any active set @xmath22 generates the same linear subspace @xmath237 , almost everywhere in @xmath13 .
in fact , we showed that for almost every @xmath13 , and for any active set @xmath22 of a solution at @xmath13 , the lasso fit can be written as @xmath399 for all @xmath218 in a neighborhood of @xmath13 , where @xmath400 is a constant ( it does not depend on @xmath218 ) .
this draws an interesting connection to linear regression , as it shows that locally the lasso fit is just a translation of the linear regression fit of on @xmath401 . the same results ( on degrees of freedom and local representations of the fit )
hold when the active set @xmath22 is replaced by the equicorrelation set @xmath179 .
our results also extend to the generalized lasso problem , with an arbitrary predictor matrix @xmath0 and arbitrary penalty matrix @xmath1 .
we showed that degrees of freedom of the generalized lasso fit is @xmath394 $ ] , with @xmath18 being the active set of any generalized lasso solution at @xmath13 , that is , @xmath402 .
as before , this result is well defined because any choice of active set @xmath22 generates the same linear subspace @xmath403 , almost everywhere in @xmath13 .
furthermore , for almost every @xmath13 , and for any active set of a solution at @xmath13 , the generalized lasso fit satisfies @xmath404 for all @xmath218 in a neighborhood of @xmath13 , where @xmath400 is a constant ( not depending on @xmath13 ) .
this again reveals an interesting connection to linear regression , since it says that locally the generalized lasso fit is a translation of the linear regression fit on @xmath0 , with the coefficients @xmath405 subject to @xmath406 .
the same statements hold with the active set @xmath22 replaced by the boundary set @xmath30 of an optimal subgradient .
we note that our results provide practically useful estimates of degrees of freedom . for the lasso problem
, we can use @xmath119 as an unbiased estimate of degrees of freedom , with @xmath22 being the active set of a lasso solution . to emphasize
what has already been said , here we can actually choose any active set ( i.e. , any solution ) , because all active sets give rise to the same @xmath119 , except for @xmath13 in a set of measure zero .
this is important , since different algorithms for the lasso can produce different solutions with different active sets . for the generalized lasso problem , an unbiased estimate for degrees of freedom
is given by @xmath407 , where @xmath22 is the active set of a generalized lasso solution .
this estimate is the same , regardless of the choice of active set ( i.e. , choice of solution ) , for almost every @xmath13 .
hence any algorithm can be used to compute a solution .
first , we prove the statement for the projection map . note that @xmath409 where the first inequality follows from ( [ eq : pfact ] ) , and the second is by cauchy
dividing both sides by @xmath410 gives the result .
we have shown that @xmath141 and @xmath142 are lipschitz ( with constant @xmath289 ) ; they are therefore continuous , and almost differentiability follows from the standard proof of the fact that a lipschitz function is differentiable almost everywhere .
we write @xmath413 to denote the set of faces of @xmath56 . to each face
@xmath414 , there is an associated normal cone @xmath415 , defined as @xmath416 the normal cone of @xmath417 satisfies @xmath418 for any @xmath419 .
[ we use @xmath420 to denote the relative interior of a set @xmath47 , and @xmath421 to denote its relative boundary . ]
now let @xmath427 .
we have @xmath428 for some @xmath424 , and by construction @xmath429 .
furthermore , we claim that projecting @xmath428 onto @xmath56 is the same as projecting @xmath11 onto the affine hull of @xmath417 , that is , @xmath430 .
otherwise there is some @xmath431 with @xmath432 , and as @xmath433 , this means that @xmath434 . by definition of @xmath435
, there is some @xmath436 such that @xmath437 . but
@xmath438 , which is a contradiction .
this proves the claim , and writing @xmath439 , we have @xmath440 as desired . in the first type of points above , vertices are excluded because @xmath448 when @xmath417 is a vertex . in the second type , @xmath56 is excluded because @xmath449 .
the lattice structure of @xmath413 tells us that for any face @xmath424 , we can write @xmath450 . this , and the fact that the normal cones have the opposite partial ordering as the faces , imply that points of the first type above can be written as @xmath451 with @xmath452 and @xmath453 for some @xmath454 . note
that actually we must have @xmath455 because otherwise we would have @xmath456 .
therefore it suffices to consider points of the second type alone , and @xmath442 can be written as @xmath457 as @xmath56 is a polyhedron , the set @xmath413 of its faces is finite , and @xmath458 for each @xmath459 .
therefore @xmath442 is a finite union of sets of dimension @xmath460 , and hence has measure zero . now let @xmath465^\perp } [ ( x_{\mathcal{e}})^+]_{(-{{\mathcal{a}}},\cdot ) } \bigl(z - ( x_{\mathcal{e}}{^t})^+\lambda s \bigr ) = 0 \bigr\}.\ ] ] the first union is taken over all possible subsets @xmath466 and all sign vectors @xmath467 ; as for the second union , we define for a fixed subset @xmath179 @xmath468^\perp } [ ( x_{\mathcal{e}})^+]_{(-{{\mathcal{a}}},\cdot ) } \not= 0 \bigr\}.\ ] ] notice that @xmath275 is a finite union of affine subspace of dimension @xmath460 , and hence has measure zero .
let @xmath252 , and let @xmath253 be a lasso solution , abbreviating @xmath18 and @xmath256 for the active set and active signs . also write @xmath212 and @xmath213 for the equicorrelation set and equicorrelation signs of the fit .
we know from ( [ eq : lassosol ] ) that we can write @xmath469 where @xmath191 is such that @xmath470_{(-{{\mathcal{a}}},\cdot ) } \bigl(y - ( x_{\mathcal{e}}{^t})^+ \lambda s \bigr ) + b_{-{{\mathcal{a } } } } = 0.\ ] ] in other words , @xmath471_{(-{{\mathcal{a}}},\cdot ) } \bigl(y - ( x_{\mathcal{e}}{^t})^+ \lambda s \bigr ) = -b_{-{{\mathcal{a } } } } \in \pi_{-{{\mathcal{a } } } } ( { \operatorname{null}}(x_{\mathcal{e}})),\ ] ] so projecting onto the orthogonal complement of the linear subspace@xmath472 gives zero , @xmath473^\perp } [ ( x_{\mathcal{e}})^+]_{(-{{\mathcal{a}}},\cdot ) } \bigl(y - ( x_{\mathcal{e}}{^t})^+\lambda s \bigr ) = 0.\ ] ] since @xmath252 , we know that @xmath473^\perp } [ ( x_{\mathcal{e}})^+]_{(-{{\mathcal{a}}},\cdot ) } = 0,\ ] ] and finally , this can be rewritten as @xmath474_{(-{{\mathcal{a}}},\cdot ) } \bigr ) \subseteq\pi_{-{{\mathcal{a}}}}({\operatorname{null}}(x_{\mathcal{e}})).\ ] ] consider defining , for a new point @xmath218 , @xmath475 where @xmath476 , and is yet to be determined .
exactly as in the proof of lemma [ lem : lcequi ] , we know that @xmath477 , and for all @xmath225 , a neighborhood of @xmath13 .
now we want to choose @xmath478 so that @xmath228 has the correct active set and active signs . for simplicity of notation , first define the function @xmath479 , @xmath480 equation ( [ eq : colsp ] ) implies that there is a @xmath476 such that @xmath481 , hence @xmath482 .
however , we must choose @xmath478 so that additionally @xmath483 for @xmath484 and @xmath485 .
write @xmath486 by the continuity of @xmath487 , there exits a neighborhood of @xmath231 of @xmath13 such that @xmath488 for @xmath484 and @xmath489 , for all @xmath233 .
therefore we only need to choose a vector @xmath476 , with @xmath481 , such that @xmath490 sufficiently small .
this can be achieved by applying the bounded inverse theorem , which says that the bijective linear map @xmath491 has a bounded inverse ( when considered a function from its row space to its column space ) .
therefore there exists some @xmath492 such that for any @xmath218 , there is a vector @xmath476 , @xmath481 , with @xmath493 finally , the continuity of @xmath494 implies that @xmath495 can be made sufficiently small by restricting @xmath496 , another neighborhood of @xmath13 .
define the set @xmath501^\perp}\cdot d_{{\mathcal{b}}\setminus{{\mathcal{a } } } } \bigl(x p_{{\operatorname{null}}(d_{-{\mathcal{b}}})}\bigr)^+ \\ & & \hspace*{165pt}{}\times \bigl(z - \bigl(p_{{\operatorname{null}}(d_{-{\mathcal{b } } } ) } x{^t}\bigr)^+ d_{\mathcal{b}}{^t}\lambda s \bigr ) = 0 \bigr\}.\end{aligned}\ ] ] the first union above is taken over all subsets @xmath502 and all sign vectors @xmath503 .
the second union is taken over subsets @xmath504 , where @xmath505^\perp } d_{{\mathcal{b}}\setminus{{\mathcal{a } } } } \bigl(x p_{{\operatorname{null}}(d_{-{\mathcal{b}}})}\bigr)^+ \not= 0 \bigr\}.\ ] ] since @xmath209 is a finite union of affine subspaces of dimension @xmath506 , it has measure zero .
now fix @xmath211 , and let @xmath354 be an optimal pair , with boundary set @xmath28 , boundary signs @xmath213 , active set @xmath18 , and active signs @xmath256 . starting from ( [ eq : genlassokktb ] ) , and plugging in for the fit in terms of @xmath352 , as in ( [ eq : genlassofit ] ) we can show that @xmath507 where @xmath508 . by ( [ eq : genlassosol ] ) , we know that @xmath509 where @xmath510 .
furthermore , @xmath511 or equivalently , @xmath512 projecting onto the orthogonal complement of the linear subspace@xmath513 therefore gives zero , @xmath514^\perp } d_{{\mathcal{b}}\setminus{{\mathcal{a } } } } \bigl(x p_{{\operatorname{null}}(d_{-{\mathcal{b}}})}\bigr)^+ \bigl(y - \bigl(p_{{\operatorname{null}}(d_{-{\mathcal{b } } } ) } x{^t}\bigr)^+ d_{\mathcal{b}}{^t}\lambda s \bigr ) = 0,\ ] ] and because @xmath211 , we know that in fact @xmath514^\perp } d_{{\mathcal{b}}\setminus{{\mathcal{a } } } } \bigl(x p_{{\operatorname{null}}(d_{-{\mathcal{b}}})}\bigr)^+ = 0.\ ] ] this can be rewritten as @xmath515 at a new point @xmath218 , consider defining @xmath516 , @xmath517 and @xmath518 where @xmath519 is yet to be determined . by construction ,
@xmath520 and @xmath228 satisfy the stationarity condition ( [ eq : genlassokkt ] ) at @xmath218 .
hence it remains to show two parts : first , we must show that this pair satisfies the subgradient condition ( [ eq : genlassosg ] ) at @xmath218 ; second , we must show this pair has boundary set , boundary signs @xmath521 , active set @xmath499 and active signs @xmath522 .
actually , it suffices to show the second part alone , because the first part is then implied by the fact that @xmath355 and @xmath253 satisfy the subgradient condition at @xmath13 .
well , by the continuity of the function @xmath523 , @xmath524 we have @xmath525 provided that @xmath225 , a neighborhood of @xmath13 .
this ensures that @xmath520 has boundary set @xmath526 and signs @xmath521 . as for the active set and signs of @xmath228 , note first that @xmath527 , following directly from the definition .
next , define the function @xmath528 , @xmath529 so @xmath530 .
equation ( [ eq : colsp2 ] ) implies that there is a vector @xmath519 such that @xmath531 , which makes @xmath532 .
however , we still need to choose @xmath478 such that @xmath533 for all @xmath484 and @xmath534 .
to this end , write @xmath535 the continuity of @xmath536 implies that there is a neighborhood @xmath231 of @xmath13 such that @xmath537 for all @xmath484 and @xmath538 , for @xmath233 . since @xmath539 where @xmath540 is the operator norm of the @xmath541
, we only need to choose @xmath519 such that @xmath531 , and such that @xmath490 is sufficiently small .
this is possible by the bounded inverse theorem applied to the linear map @xmath542 : when considered a function from its row space to its column space , @xmath542 is bijective and hence has a bounded inverse .
therefore there is some @xmath492 such that for any @xmath218 , there is a @xmath519 with @xmath531 and @xmath543 the continuity of @xmath544 implies that the right - hand side above can be made sufficiently small by restricting @xmath496 , a neighborhood of @xmath13 .
the dual of the lasso problem ( [ eq : lasso ] ) has appeared in many papers in the literature ; as far as we can tell , it was first considered by @xcite .
we start by rewriting problem ( [ eq : lasso ] ) as @xmath545 then we write the lagrangian @xmath546 and we minimize @xmath547 over @xmath548 to obtain the dual problem @xmath549 taking the gradient of @xmath547 with respect to to @xmath548 , and setting this equal to zero gives @xmath550 where @xmath160 is a subgradient of the function @xmath551 evaluated at @xmath162 . from ( [ eq : lassodual1 ] ) , we can immediately see that the dual solution @xmath552 is the projection of @xmath13 onto the polyhedron @xmath56 as in lemma [ lem : lassoproj ] , and then ( [ eq : lassopd1 ] ) shows that @xmath553 is the residual from projecting @xmath13 onto @xmath56 .
further , from ( [ eq : lassopd2 ] ) , we can define the equicorrelation set @xmath179 as @xmath554 noting that together ( [ eq : lassopd1 ] ) , ( [ eq : lassopd2 ] ) are exactly the same as the kkt conditions ( [ eq : lassokkt ] ) , ( [ eq : lassosg ] ) , and all of the arguments in section [ sec : lasso ] involving the equicorrelation set @xmath179 can be translated to this dual perspective .
there is a slightly different way to derive the lasso dual , resulting in a different ( but of course , equivalent ) formulation .
we first rewrite problem ( [ eq : lasso ] ) as @xmath555 and by following similar steps to those above , we arrive at the dual problem @xmath556 each dual solution @xmath552 ( now no longer unique ) satisfies @xmath557 the dual problem ( [ eq : lassodual2 ] ) and its relationship ( [ eq : lassopd3 ] ) , ( [ eq : lassopd4 ] ) to the primal problem offer yet another viewpoint to understand some of the results in section [ sec : lasso ] . for the generalized lasso problem
, one might imagine that there are three different dual problems , corresponding to the three different ways of introducing an auxiliary variable @xmath558 into the generalized lasso criterion : @xmath559 { { \hat{\beta}}},\hat{z } & \in&\mathop{{\operatorname{argmin}}}_{\beta\in{\mathbb{r}}^p , z \in{\mathbb{r}}^p } { \frac{1}{2}}\|y - x\beta\|_{2}^2 + \lambda\|dz\|_{1}\qquad\mbox{subject to } z=\beta ; \\[2pt ] { { \hat{\beta}}},\hat{z } & \in&\mathop{{\operatorname{argmin}}}_{\beta\in{\mathbb{r}}^p , z \in{\mathbb{r}}^m } { \frac{1}{2}}\|y - x\beta\|_{2}^2 + \lambda\|z\|_{1}\qquad\mbox{subject to } z = d\beta.\end{aligned}\ ] ] however , the first two approaches above lead to lagrangian functions that can not be minimized analytically over @xmath548 .
only the third approach yields a dual problem in closed - form , as given by @xcite , @xmath560 \\[-7pt ] \eqntext{\mbox{subject to } \|v\|_{\infty}\leq\lambda , d{^t}v \in{\operatorname{row}}(x).}\end{aligned}\ ] ] the relationship between primal and dual solutions is @xmath561 \label{eq : genlassopd2 } { \hat{v}}&= & \lambda\gamma,\end{aligned}\ ] ] where @xmath300 is a subgradient of @xmath301 evaluated at @xmath302 . directly from ( [ eq : genlassodual ] ) we can see that @xmath562 is the projection of the point @xmath563 onto the polyhedron @xmath564 by ( [ eq : genlassopd1 ] ) , the primal fit is @xmath565 , which can be rewritten as @xmath566 where @xmath56 is the polyhedron from lemma [ lem : genlassoproj ] , and finally @xmath567 because @xmath142 is zero on @xmath568 . by ( [ eq : genlassopd2 ] ) , we can define the boundary set @xmath30 corresponding to a particular dual solution @xmath552 as @xmath569 ( this explains its name , as @xmath30 gives the coordinates of @xmath552 that are on the boundary of the box @xmath570 . ) as ( [ eq : genlassopd1 ] ) , ( [ eq : genlassopd2 ] ) are equivalent to the kkt conditions ( [ eq : genlassokkt ] ) , ( [ eq : genlassosg ] ) [ following from rewriting ( [ eq : genlassopd1 ] ) using @xmath571 , the results in section [ sec : genlasso ] on the boundary set @xmath30 can all be derived from this dual setting . | we derive the degrees of freedom of the lasso fit , placing no assumptions on the predictor matrix @xmath0 .
like the well - known result of zou , hastie and tibshirani [ _ ann .
statist . _
* 35 * ( 2007 ) 21732192 ] , which gives the degrees of freedom of the lasso fit when @xmath0 has full column rank , we express our result in terms of the active set of a lasso solution .
we extend this result to cover the degrees of freedom of the generalized lasso fit for an arbitrary predictor matrix @xmath0 ( and an arbitrary penalty matrix @xmath1 ) . though our focus is degrees of freedom , we establish some intermediate results on the lasso and generalized lasso that may be interesting on their own . . |
the final state interaction ( fsi ) in the nonleptonic @xmath0 decay has been an important unsolved issue in connection with the search of direct cp violations . unlike the short - distance fsi
, the long - distance fsi has not been understood well enough even qualitatively .
the experimental data of the @xmath5 decay clearly show that the fsi phases are large in the @xmath6 decay modes@xcite .
opinions divide as to how strong the fsi is in the @xmath0 decay .
some theorists have suggested that the long - distance fsi should be small at the mass scale of the @xmath0 meson .
but others have obtained large fsi phases by numerical computations based on various dynamical assumptions and approximations . according to the latest data ,
the fsi phases are tightly bounded for @xmath7 and a little less so for @xmath8 , @xmath9 and @xmath10@xcite .
however , the tight bounds are closely tied to smallness of the so - called color - suppressed modes .
is the smallness of the fsi phases special only to those sets of modes for which the color suppression occurs ? if it is more general , where does transition occur from large fsi phases to small fsi phases in terms of the mass scale of a decaying particle ?
although the process is not a weak decay , the @xmath2 decay falls between the @xmath5 decay and the @xmath0 decay in terms of energy scale . since the time scale of the strong and electromagnetic decay processes of @xmath2 is much shorter than that of the long - distance fsi , the decay interactions of @xmath2 act just like the weak interactions of the @xmath5 and the @xmath0 decay as far as the long - distance fsi is concerned . for this reason ,
analysis of the @xmath2 decay amplitudes provides one extrapolation from the @xmath5 mass toward the @xmath0 mass . among the two - body decay modes of @xmath2 , most extensively measured are the @xmath11 modes .
a detailed analysis of those decay amplitudes with broken flavor su(3 ) symmetry found a large relative phase of fsi ( @xmath12 ) between the one - photon and the gluon decay amplitudes@xcite .
since there are many independent su(3 ) amplitudes for the @xmath11 decay , the analysis involved one assumption of simplification on assignment of the fsi phases . in this short paper , we shall study the @xmath1 decay modes of @xmath2 which are much simpler in the su(3 ) structure .
the result of analysis turns out clearer and more convincing .
once the asymptotic behavior of the electromagnetic form factors is incorporated in analysis , the current data favor a very large fsi phase difference between the one - photon and the gluon decay amplitudes .
in order to formulate the fsi , it is customary to separate interactions into three parts , the decay interaction , the rescattering interaction , and the hadron formation interaction .
separation between the second and the third can be done only heuristically at best , not at the level of lagrangian .
one way to circumvent this ambiguity and see general properties of the fsi is to break up decay amplitudes in the eigenchannels of the strong interaction s - matrix : @xmath13 an observed two - body final state can be expanded in the eigenchannels with an orthogonal matrix as @xmath14 where the superscript `` in '' stands for the incoming state . in terms of the `` in '' and `` out '' states , the s - matrix of eq.([eigenchannel ] ) can be expressed as @xmath15 .
when the effective decay interactions @xmath16 , in which we include all coefficients , are time - reversal invariant , the decay amplitude for @xmath17 is given in the form @xmath18 where @xmath19 is the decay amplitude into the eigenchannel @xmath20 through @xmath16 ; @xmath21 and @xmath22 is real . contains a short - distance fsi phase , which is transferred into @xmath23 in eq.([6 ] ) . ]
two interactions are relevant to the @xmath2 decay .
for the one - photon annihilation , @xmath24 , where @xmath25 is the vector field of @xmath2 .
for the gluon annihilation , @xmath26 where @xmath27 is a vector function of the gluon field tensor @xmath28 and its derivatives which is calculated in perturbative qcd . when the terms from the same decay interaction are grouped together , eq.([general ] ) takes the form , @xmath29 where @xmath30 we emphasize here that the net fsi phase @xmath23 of @xmath31 depends on @xmath16 through @xmath32 even for the same state @xmath33 when more than one eigenchannel is open . specifically in the @xmath2 decay
, @xmath23 is different between the one - photon and the three - gluon decay amplitude even for the same isospin state .
if the fsi is strong in the @xmath2 decay , a large phase difference @xmath34 can arise .
our aim is to learn about @xmath35 from the decay @xmath36 .
one feature of the @xmath37 is particularly advantageous to our study : there is no su(3 ) symmetric decay amplitude for the gluon decay .
charge conjugation does not allow a @xmath1 state to be in an su(3 ) singlet state of @xmath38 .
therefore the @xmath1 final states through the gluon decay must be in an octet along the su(3 ) breaking direction of @xmath39 .
since the leading term of the three - gluon decay is su(3)-breaking , the one - photon process competes with the otherwise dominant gluon process , making it easier to determine a relative fsi phase through interference .
the @xmath2 amplitudes are parametrized in terms of the reduced su(3 ) amplitudes , @xmath40 , @xmath41 , and @xmath42 , as follows : @xmath43 where @xmath44 and @xmath45 are the @xmath46 flavor matrices of the @xmath47meson octet and @xmath48)/2 .
@xmath40 is for the gluon decay while @xmath41 and @xmath42 are for the one - photon annihilation and the su(3 ) breaking correction to it , respectively .
breaking to the one - photon annihilation @xmath49 has the same group structure as @xmath40 . ]
no * 10 * or @xmath50 representation of @xmath1 arises from multiple insertions of @xmath39 alone .
charge conjugation invariance amounts to antisymmetrization in @xmath51 , which forbids the * 27 * -representation of @xmath1 .
we have normalized each reduced amplitude such that sum of individual amplitudes squared be common .
the decay amplitudes for the observed modes are listed in table 1 in this parametrization .
also listed are the absolute values of the measured amplitudes@xcite after small phase - space corrections are made .
if the flavor su(3 ) is a decent symmetry , @xmath42 must be a fraction of @xmath41 .
knowing the magnitude of typical flavor - su(3 ) breakings , let us allow @xmath52
the one - photon annihilation amplitudes @xmath41 and @xmath42 describe the electromagnetic form factors too .
we have some theoretical understanding of their asymptotic behaviors . according to the perturbative qcd analysis@xcite
, the leading asymptotic behavior of the form factor for meson @xmath53 is given by @xmath54^{\gamma_i } } \biggr ) , \label{asym}\ ] ] where @xmath55 is the decay constant of @xmath56 , @xmath57 is the qcd coupling , and @xmath58 are positive constants .
@xmath59 approaches a real value as @xmath60 .
therefore , the one - photon amplitudes @xmath41 and @xmath42 have a common phase ( = 0 ) in this limit : @xmath61 since @xmath62 , we expect @xmath63 .
the physical picture of this inequality is obvious in the spacelike region of @xmath64 : difference between @xmath65 and @xmath66 is @xmath67 _ vs _ @xmath68 in the valence quark content .
the @xmath68 quark , being a little heavier , is harder in momentum distribution inside @xmath65 than the @xmath67 quark is inside @xmath66 .
this leads to a stiffer form factor for @xmath65 than for @xmath66 , though an accurate theoretical estimate of @xmath69 is not possible for finite @xmath64 . with @xmath70 , @xmath71 as @xmath72 . combining eq.([ratio ] ) with eq.([arg ] ) , we obtain @xmath73 .
when we keep the nonleading logarithmic terms , there is a small relative phase between @xmath41 and @xmath42 ; @xmath74^{1+\gamma_1}}\biggr).\ ] ] we may ignore it since it can be treated as a correction to the symmetry - breaking correction term @xmath42 .
if it happens that vector resonances of light quarks exist just around the @xmath2 mass , the form factors would not be asymptotic at this energy . if such high mass resonances should have a substantial branching into the @xmath1 channels
, the nonleading logarithmic terms would add up to a nonnegligible magnitude in eq.([asym ] ) . in this case the phases of @xmath41 and @xmath42 would not be small .
one may wonder about whether a mass splitting of the resonances might generate a large phase difference between @xmath42 and @xmath41 .
however , the widths of such resonances , if any , would be so broad at such high mass that the mass splitting effect would be largely washed out . and @xmath41 .
] therefore we expect that the phase equality of eq.([arg ] ) should hold in a good approximation . in our numerical analysis
we shall set the phases of @xmath41 and @xmath42 to a common value and impose the condition of @xmath75 at @xmath76 : @xmath77 if we attempt to fit the data with the leading terms @xmath40 and @xmath41 alone without fsi phases , the result is unacceptable .
the fit of the minimum @xmath78 is obtained for @xmath79 and @xmath80 leading to @xmath81 for only three data .
we then include @xmath42 to fit the data . if we ignored the constraint of eq.([constraint ] ) , the amplitudes could be fitted with @xmath82 this set of numbers would give @xmath83 contrary to @xmath84 .
when we include the constraint @xmath85 , the fit of the best @xmath78 is back to @xmath86 of @xmath81 . the same poor fit with @xmath40 and @xmath41 alone .
it is fairly obvious why we can not fit the data .
looking up the parametrization in table i , we see that without phases the @xmath87 amplitude must be larger in magnitude than sum of the @xmath88 and the @xmath89 amplitude for @xmath90 .
the measured values badly violate this inequality .
the natural recourse is to introduce fsi phases for the amplitudes .
we first try with @xmath40 and @xmath41 alone . defining the relative fsi phase between @xmath40 and @xmath41 by @xmath91 we can fit the amplitudes with @xmath92 where @xmath35 is defined between @xmath93 and @xmath94 .
the attached uncertainty comes from the experimental errors of the branching fractions , which are treated as uncorrelated here . since we determine @xmath35 through @xmath95 which is sensitive to small experimental errors near @xmath96 , the uncertainty in eq.([thefit ] ) turns out to be a little larger than one might expect from those of the branching fractions .
one may wonder how much @xmath35 can be reduced by adding the breaking term @xmath42 with the constraint of eq.([constraint ] ) .
the result is plotted in fig.1 .
dependence of @xmath35 on the ratio @xmath97 is very mild : @xmath35 decreases slowly and monotonically from @xmath98 at @xmath99 to @xmath100 at the edge of the allowed range , @xmath101 .
even if the fsi phases of @xmath41 and @xmath42 are left independent , it is fairly obvious that we can not fit the data with small values for all phase differences under the constraint @xmath102 .
we have thus come to the conclusion that the fsi phase difference between the one - photon and the gluon decay amplitudes is very large , as large as @xmath98 . for this magnitude
, it must come mostly from the long - distance fsi .
our conclusion of large fsi phases has a profound implication in the @xmath0 decay .
the important input leading to this conclusion is that the electromagnetic form factor of @xmath65 does not fall faster than that of @xmath66 .
while it is very reasonable in perturbative qcd , we can in principle test this postulate in experiment . just measure the ratio of the one - photon annihilation cross sections for @xmath103 and @xmath87 off the @xmath2 peak .
we do not have good data on the ratio @xmath104 off the peak .
experiment requires time and a good @xmath105 separation .
a measurement will certainly have a great impact on the issue of the long - distance fsi in heavy particle decays .
the magnitude of measured cross sections will also tell how close the form factors are to their asymptotic limits and therefore how small the phases of @xmath41 and @xmath42 are .
even a value of the unseparated ratio @xmath106 off the peak will throw in one more input in the analysis .
i am grateful to s. brodsky for an instruction in the perturbative ocd analysis of the form factors .
this work was supported in part by the director , office of energy research , office of high energy and nuclear physics , division of high energy physics of the u.s .
department of energy under contract de ac0376sf00098 and in part by the national science foundation under grant phy9514797 .
m. bishai et al.(cleo collab ) , phys .
lett . * 78 * , 3261 ( 1997 ) .
m. suzuki , phys .
rev . d*57 * , 5717 ( 1998 ) .
m. suzuki , phys .
rev . d*58 * , 111504 ( 1998 ) .
particle data group , _ review of particle physics _ ,
j. * c3 * , 1 ( 1998 ) .
farrar and d.r .
jackson , phys .
lett . * 43 * , 246 ( 1979 ) : a.v .
efremov and a.v .
radyushkin , phys . lett . *
94b * , 245 ( 1980 ) : g.p .
lepage and s. brodsky , phys . lett . * 87b * , 359 ( 1979 ) ; phys .
. lett . * 43 * , 545 ( 1979 ) ; * 43 * , 1625(e ) ( 1979 ) .
lepage and s. brodsky , phys . rev .
* 22 * , 2157 ( 1980 ) .
.the su(3 ) parametrization of the @xmath1 decay amplitudes of @xmath2 and their magnitudes from the observed decay branching fractions .
the central value of the @xmath88 amplitude is normalized to unity . [
cols="^,^,^,^",options="header " , ] | in view of important implications in the @xmath0 decay , the @xmath1 decay modes of @xmath2 are analyzed with broken flavor su(3 ) symmetry in search for long - distance final - state interactions .
if we impose one mild theoretical constraint on the electromagnetic form factors , we find that a large phase difference of final - state interactions is strongly favored between the one - photon and the gluon decay amplitudes .
measurement of @xmath3 and @xmath4 off the @xmath2 peak can settle the issue without recourse to theory . |
superconductivity in hole - doped ( ba@xmath2k@xmath3)fe@xmath4as@xmath4 @xcite ( bak122 , in the following ) was found soon after the discovery of superconductivity with high critical temperatures in oxypnictide feas - based materials @xcite .
several studies reported anisotropic properties of single crystals , representative of various parts of the phase diagram @xcite , however , no systematic evolution of the resistivity and its anisotropy was undertaken so far , due to a difficulty in preparation of single crystals with well controlled potassium content .
high quality single crystals of k - doped materials can be grown from feas flux @xcite , however , the high melting temperature of the flux , leading to high potassium vapor pressure , limits this technique to growth of only the underdoped compositions .
the compositions on the overdoped side can be grown from kas flux @xcite .
initial success in growing single crystals through the use of sn flux , and finding their low anisotropy through measurements of the upper critical field @xcite , was stopped by finding of the macroscopic phase separation in sn - flux @xcite or polycrystalline @xcite samples and gross effect of small sn incorporation on the phase diagram @xcite .
measurements of anisotropy are of great importance for understanding the normal state of iron pnictide superconductors .
for example , careful characterization of the anisotropic resistivity in electron - doped ba(fe@xmath2co@xmath3)@xmath4as@xmath4 ( baco122 in the following ) found unusual anisotropy of transport both in the normal @xcite and in the superconducting @xcite states .
the in - plane transport reveals close to linear @xmath13 at optimal doping , which evolves systematically towards @xmath14 behavior in the heavily overdoped compositions , suggestive of a quantum critical point at optimal doping @xcite .
the inter - plane resistivity , @xmath15 , on the other hand , reveals an unusual maximum @xcite , correlating well with temperature - dependent nmr knight shift @xcite representative of a pseudogap .
the characteristic temperature of the pseudogap decreases with co - doping and vanishes at @xmath16=0.31 recovering normal metallic properties , in particular suppressing temperature - dependent spin susceptibility @xcite and hall effect @xcite .
broad crossover with notable slope change of temperature - dependent resistivity is also observed in in - plane transport in single crystals of bak122 at doping close to optimal @xcite , similar to pure stoichiometric kfe@xmath4as@xmath4 ( k122 ) @xcite .
it was suggested that this unusual temperature dependence of resistivity stems from multi - band effects @xcite , with the contribution of conductivity channels with nearly temperature - independent and strongly temperature - dependent resistivities . on the other hand , multi - component analysis of in - plane resistivity @xmath17 in isoelectron ru - doped baru122
, suggests that a crossover - type temperature dependence is characteristic of hole contribution , while electronic contribution is close to @xmath18-linear @xcite .
in this article we perform detailed study of the temperature - dependent in - plane and inter - plane resistivity of bak122 over a broad doping range from parent compound to close to optimal doping level @xmath19 @xcite .
we show that the unusual temperature dependence of the in - plane resistivity correlates with the pseudo - gap resistivity maximum in the inter - plane resistivity .
this is dramatically different from the lack of any pseudo - gap related features in the temperature - dependent in - plane resistivity of electron - doped materials .
another difference between electron- and hole- doped materials is an increase of the resistivity crossover temperature @xmath20 in bak122 with doping .
single crystals of bak122 were grown using high temperature feas flux technique @xcite . because of the volatility of k during growth , single crystals have a distribution of potassium content , with inner parts of the crystals frequently having @xmath21 differing by 1 to 3 k from the surface parts . for our study
we selected samples using the sharpness of the superconducting transition as a measure of constant dopant concentration .
samples for the study were cut from the inner parts of single crystals . after cutting , we performed precision magnetic susceptibility measurements so that we could inspect the samples for possible inclusions with lower @xmath21 .
in addition samples were extensively characterized by magneto - optic techniques to look for possible inhomogeneity , as described in detail in ref . .
only samples with sharp transitions were selected .
the chemical composition was measured on selected crystals using wavelength dispersive x - ray spectroscopy ( wds ) in jeol jxa-8200 electron microprobe .
the composition was measured for 12 points per single crystal and averaged .
samples for in - plane resistivity measurements had typical dimensions of ( 1- 2)@xmath220.5@xmath22(0.02 - 0.1 ) mm@xmath23 .
all sample dimensions were measured with an accuracy of about 10% .
contacts for four - probe resistivity measurements were made by soldering 50 @xmath24 m silver wires with ultrapure sn solder , as described in ref . .
this technique produced contact resistance typically in the 10 @xmath25 range .
inter - plane resistivity was measured using a two - probe technique , relying on the negligibly small contact resistance .
samples typically had dimensions [email protected]@xmath220.1 mm@xmath23 ( @xmath26 ) , their top and bottom ab - plane surfaces were covered with sn solder forming a capacitor - like structure .
four - probe scheme was used down to the sample to measure series connected sample , @xmath27 , and contact , @xmath28 resistance .
taking into account that @xmath29 , contact resistance represents a minor correction of the order of 1 to 5% .
this can be directly seen for our samples for temperatures below the superconducting @xmath21 , where @xmath300 and the measured resistance represents @xmath28 @xcite .
the details of the measurement procedure can be found in refs . .
the drawback of the measurement on samples with @xmath31 is that any inhomogeneity in the contact resistivity or internal sample connectivity admixes the in - plane component due to redistribution of the current . to minimize this effect
, we performed measurements of @xmath32 on at least 5 samples of each composition . in all cases we obtained qualitatively similar temperature dependencies of the electrical resistivity , as represented by the ratio of resistivities at room and low temperatures , @xmath33 .
the resistivity value , however , showed a notable scatting and at room temperature was typically in the range 1000 to 2000 @xmath25 cm .
[ f1_tcx ] shows superconducting @xmath21 as a function of potassium content .
here we compare our measurements on representative single crystals with wds k - content determination with literature data as determined from resistivity and magnetization @xcite , specific heat @xcite , neutron scattering and magnetization measurements @xcite on polycrystalline materials .
thus determined @xmath34 are in good agreement with each other and with measurements on single crystals @xcite .
however , because the samples cut from various parts of the same crystal can have variation of @xmath21 , we adopted the following approach in determining @xmath16 for each sample .
we fitted the @xmath34 determined by avci et al .
, using two functions . in the underdoped regime
we used a parabolic function , an approach frequently used in the cuprates @xcite . for underdoped compositions this resulted in the formula @xmath35 , as shown by solid line in fig .
[ f1_tcx ] .
as can be seen from the comparison of the curve with actual data , the curve fits nicely the rapid rise of @xmath21 in the underdoped regime , the main interest in our study .
close to optimal doping the fit is not as good , the curve fails to reproduce the rather flat dependence in @xmath16 range from 0.3 to 0.5 @xcite . for overdoped compositions
the @xmath36 is close to linear . as determined in this study , with @xmath21 determined from zero resistance criterion and @xmath37 from wds measurements for the representative crystals .
solid black line shows a fit of the @xmath36 used to determine @xmath16 for individual crystals in this and related @xcite studies .
dashed purple line is a fit of @xmath36 in the overdoped regime .
, width=302 ] fig .
[ f2_inplane ] shows temperature - dependent in - plane resistivity of bak122 for several compositions from parent , @xmath16=0 , to close to optimally doped , @xmath16=0.35 .
several features should be noticed .
first , the curves for samples with @xmath38 ( @xmath39 26k ) show a small but clear anomaly on passing through the temperature of coupled structural / magnetic transition @xmath7 , with resistivity accelerating its decrease on cooling .
the minimum distinct value of @xmath7=70 k is observed in samples with @xmath21=26 k , and we could not resolve any feature in samples with @xmath40 .
both these observations are in good agreement with the position of the concentration boundary in neutron scattering data @xcite .
in addition to a feature at @xmath7 , the in - plane resistivity shows a broad crossover with a slope change at a temperature of approximately 200 k , similar to previous measurements by golubov et al .
similar crossover at about the same temperature is seen even in the terminal composition kfe@xmath4as@xmath4 @xcite and in polycrystalline samples @xcite . ,
normalized to room temperature values , @xmath41 .
the curves are offset to avoid overlapping .
data for kfe@xmath4as@xmath4 are from ref . .
down - triangles show the position of the maximum of the inter - plane resistivity.,width=302 ] it is important to notice that the crossover feature is observed through all compositions from non - superconducting @xmath16=0.12 to heavily overdoped kfe@xmath4as@xmath4 ( @xmath16=1 ) , without any significant shift on a temperature scale .
this insensitivity to doping is difficult to explain in any multi - band model , and by comparison with inter - plane resistivity , we suggest that this resistivity feature is a reflection of a pseudogap crossover at @xmath20 in the inter - plane transport , whose position is shown by arrows in fig .
[ f2_inplane ] .
another interesting observation is the very weak , if any , dependence of the value of room temperature resistivity on @xmath16 .
for all compositions the resistivity stayed close to 300 @xmath25 cm .
this compares favorably to the values reported for kfe@xmath4as@xmath4 : 280 @xmath25 cm @xcite , 360 @xmath25 cm @xcite , 480 @xmath25 cm @xcite and 600 @xmath25 cm @xcite , with scatter presumably determined by cracks @xcite and inaccuracy of geometric factor determination .
, normalized to room temperature values , @xmath42 .
the curves are offset to avoid overlapping .
the data for kfe@xmath4as@xmath4 ( @xmath16=1 ) are from ref . .
down triangles show position of the @xmath43 maximum , plotted in fig .
[ f2_inplane ] and in the phase diagram fig .
[ f5_pseudogap ] . ,
width=302 ] ( arrows ) in a doping range close to @xmath440.4 .
the curves are offset to avoid overlapping .
a small increase of the derivative preceding @xmath21 in samples @xmath16=0.28 and @xmath16=0.35 shifts up with the superconducting transition temperature and is most likely due to superconducting fluctuations .
down triangles show positions where the derivative is zero corresponding to the maximum of the inter - plane resistivity .
, width=302 ] in fig .
[ f3_interplane ] we plot the temperature - dependent inter - plane resistivity of bak122 .
the feature due to the structural transition is much clearer in the @xmath1 , and we determine its position by plotting the derivative of the @xmath43 , as shown in fig . [ f4_interplanederivative ] .
in addition to the sharp feature in derivative in sample with @xmath16=0.25 at @xmath7=70 k , a slight increase of the resistivity derivative is seen in samples with @xmath16=0.28 and @xmath16=0.35 .
this feature shifts to higher temperature with increasing superconducting @xmath21 and is most likely due to superconducting fluctuations .
note , that beyond this feature , samples with @xmath16=0.28 and @xmath16=0.35 , close to optimal doping @xmath45=0.4 , show nearly linear temperature - dependent resistivity . this can be seen as a flattening of the resistivity derivative below @xmath46100 k above the superconducting transition .
all these features in the temperature - dependent inter - plane resistivity are seen much below the temperature of the crossover point of resistivity from non - metallic behavior at high temperatures to metallic , with @xmath1 decreasing on cooling , at low temperatures .
the crossover temperature , @xmath20 , shifts up very slightly with doping , from @xmath46200 k in parent compound to @xmath46250 k for bak122 with @xmath16=0.35 ( close to optimal doping ) .
it is interesting that @xmath20 shows a strongly asymmetric doping dependence , with a rapid decrease in electron - doped baco122 and a slow rise in hole- doped bak122 , as summarized in the phase diagram plot of fig .
[ f5_pseudogap ] .
another distinct feature between the resistivity behavior of electron- and hole - doped materials is that the pseudogap affects both in - plane and inter - plane transport in hole - doped materials .
the effect of @xmath20 on the in - plane transport is negligible in electron - doped materials .
the superconducting @xmath21 was determined from measurements on polycrystals ( open circles ) , refs . ,
see fig .
[ f1_tcx ] .
dark - yellow solid squares show the data for single crystals , ref .
, red stars show the @xmath36 as determined in this study .
the temperatures of the structural(solid blue dots ) and magnetic ( open blue stars ) transitions are from neutron scattering , ref . and x - ray ( solid dots ) study , ref .
, red up - triangles show positions of the corresponding features in the inter - plane transport , fig .
[ f3_interplane ] .
red down - triangles show positions of maxima in the inter - plane resistivity at @xmath20 . for comparison
we show the phase diagram for electron - doped ba(fe@xmath2co@xmath3)@xmath4as@xmath4 , ref . .
green and magenta lines represent positions of the maximum , @xmath20 , and the minimum , @xmath47 , in temperature - dependent inter - plane resistivity @xcite .
the orange line is guide to the eyes .
, width=302 ]
contrary to baco122 , magnetic ordering and the structural transition happen simultaneously in bak122 at a temperature @xmath48=@xmath49 @xcite .
magnetic ordering should reconstruct the fermi surface , opening a nesting or superzone gaps on electron and hole pockets . in hole - doped materials
this gap opening , instead of leading to a resistivity increase , leads to an accelerated resistivity decrease ( increase of resistivity derivative ) , suggesting that the main effect comes from a change in the inelastic scattering due to taming down the contribution of pre - transition fluctuations of the order parameter .
the parts of the fermi surface which are not affected by the sdw gap @xcite , enjoy a notably reduced inelastic scattering in the magnetically ordered phase @xcite .
in the parent compound , in which disorder , introduced by dopants , is absent and thus elastic scattering is small , this decrease of scattering overcomes the loss of the carrier density so that the total conductivity increases below @xmath7 . since the inter - plane transport is dominated by the most warped parts of the fermi surface @xcite , least affected by the sdw super - zone gap , the inter - plane resistivity should be affected much less by the sdw gap opening than @xmath50 .
this is indeed seen , in bak122 , very similar to baco122
. in baco122 the decrease of the inter - plane resistivity below @xmath20 shows a clear correlation with the nmr knight shift , which was the reason for our suggestion of its relation to a pseudogap . at temperatures below @xmath20
both the knight shift and the inter - plane resistivity in baco122 follow the expectations of a metal with temperature - independent density of states , but become temperature - dependent at @xmath51 . very recently
similar measurements have been undertaken in optimally doped bak122 @xcite , and in stoichiometric kfe@xmath4as@xmath4 @xcite . at optimal
doping the nmr knight shift is increasing with temperature while the spin - relaxation rate , 1/@xmath52 , decreases and becomes constant above @xmath46200 k. in overdoped kfe@xmath4as@xmath4 the knight shift was found to be nearly temperature independent , however , the temperature dependence of the korringa ratio indicated strong electron correlations . we need to notice though , that the decrease of the inter - plane resistivity , despite being very small , would be very difficult to explain by only a change of the scattering mechanism
. it would require activation of carriers by excitations over the partial gap on the fermi surface ( pseudogap ) .
measurements of the inter - plane resistivity in bak122 show that the magnetic / structural transition does not lead to resistivity increase , i.e. the loss of the carrier density at the transition is insignificant . on transition suppression with doping , temperature - dependent inter - plane resistivity becomes nearly linear above the superconducting transition , suggesting the existence of a quantum critical point .
in addition to a feature in the temperature - dependent resistivity at @xmath7 , a broad maximum is found at a temperature @xmath53200 k , which by comparison with the electron - doped baco122 @xcite , we assign to the formation of pseudogap . despite the effect of doping in multi - band metallic system
may be quite complicated , comparison of the hole - doped bak122 with the electron - doped baco122 shows significant difference .
a pseudogap resistive crossover at @xmath20 in the inter - plane resistivity vanishes with doping in baco122 but remains intact in bak122 .
the crossover affects temperature - dependent in - plane resistivity in bak122 , however , it does not in baco122 . on co - doping the in - plane resistivity at room temperature decreases by approximately a factor of 3 from @xmath16=0 to @xmath54=0.31 , while it does not show much variation on hole doping .
this result may suggest that inelastic scattering is strongly suppressed in the case of co - doping , while it remains strong for hole - doped compositions .
since a similar trend is seen for magnetic fluctuations , it is natural to expect that magnetic scattering contributes significantly to normal state inelastic scattering . supporting this conclusion
, baco122 doped beyond the superconducting dome shows simple metallic behavior in resistivity ( both in- and inter - plane ) and pauli susceptibility @xcite , and temperature - independent hall constant @xcite , with all pseudogap features suppressed by @xmath54=0.31 . on the other hand , anomalous temperature - dependent resistivity with the resistivity crossover , suggesting the existence of the pseudogap ,
is preserved in kfe@xmath4as@xmath4 .
finally we would like to point to a certain similarity in the critical behavior of the inter - plane resistivity in bak122 and in cecoin@xmath55 . in cecoin@xmath55 , a true critical behavior at a field - tuned qcp @xcite with @xmath18-linear resistivity and violation of the wiedemann - franz law
is observed for transport along the tetragonal @xmath56-axis @xcite , while transport along the plane obeys the wiedemann - franz law @xcite .
this is similar to the difference in the temperature dependence of @xmath13 and @xmath12 in bak122 , with the latter being more linear at optimal doping .
we thank t. terashima and m. kimata for sharing resistivity data files , and y. liu and c. t. lin for providing unpublished doping - dependent @xmath21 data .
work at the ames laboratory was supported by the department of energy - basic energy sciences under contract no .
de - ac02 - 07ch11358 .
work in china was supported by the ministry of science and technology of china , project 2011cba00102 .
n. ni , s. l. budko , a. kreyssig , s. nandi , g. e. rustan , a. i. goldman , s. gupta , j. d. corbett , a. kracher , p. c. canfield phys .
b * 78 * , 014507 ( 2008 ) .
u. welp , r. xie , a. e. koshelev , w. k. kwok , h. q. luo , z. s. wang , g. mu , and h. h. wen phys .
b * 79 * , 094505 ( 2009 ) .
taichi terashima , motoi kimata , hidetaka satsukawa , atsushi harada , kaori hazama , shinya uji , hisatomo harima , gen - fu chen , jian lin luo , and nan - ling wang , j. phys .
soc . jpn . *
78 * , 063702 ( 2010 ) .
k. hashimoto , a. serafin , s. tonegawa , r. katsumata , r. okazaki , t. saito , h. fukazawa , y. kohori , k. kihou , phys .
b * 82 * , 014526 ( 2010 ) .
kunihiro kihou , taku saito , shigeyuki ishida , masamichi nakajima , yasuhide tomioka , hideto fukazawa , yoh kohori , toshimitsu ito , shin - ichi uchida , akira iyo , chul - ho lee , and hiroshi eisaki , j. phys .
. jpn . * 79 * , 124713 ( 2010 ) .
a. a. aczel , e. baggio - saitovitch , s. l. budko , p.c .
canfield , j. p. carlo , g. f. chen , pengcheng dai , t. goko , w. z. hu , g. m. luke , j. l. luo , n. ni , d. r. sanchez - candela , f. f. tafti , n. l. wang , t. j. williams , w. yu , y. j. uemura phys .
b * 78 * , 214503 ( 2008 ) .
d. s. inosov , a. leineweber , xiaoping yang , j. t. park , n. b. christensen , r. dinnebier , g. l. sun , ch .
niedermayer , d. haug , p. w. stephens , j. stahn , o. khvostikova , c. t. lin , o. k. andersen , b. keimer , and v. hinkov , phys .
b * 79 * , 224503 ( 2009 ) .
t. goko , a. a. aczel , e. baggio - saitovitch , s. l. budko , p. c. canfield , j. p. carlo , g. f. chen , pengcheng dai , a. c. hamann , w. z. hu , h. kageyama , g. m. luke , j. l. luo , b. nachumi , n. ni , d. reznik , d. r. sanchez - candela , a. t. savici , k. j. sikes , n. l. wang , c. r. wiebe , t. j. williams , t. yamamoto , w. yu , and y. j. uemura , phys .
b * 80 * , 024508 ( 2008 ) .
hideto fukazawa , takehiro yamazaki , kenji kondo , yoh kohori , nao takeshita , parasharam m. shirage , kunihiro kihou , kiichi miyazawa , hijiri kito , hiroshi eisaki , and akira iyo , j. phys .
. jpn . * 78 * , 033704 ( 2010 ) .
m. a. tanatar , n. ni , c. martin , r. t. gordon , h. kim , v. g. kogan , g. d. samolyuk , s. l. budko , p. c. canfield , and r. prozorov , phys .
b * 79 * , 094507 ( 2009 ) .
m. a. tanatar , n. ni , g. d. samolyuk , s. l. budko , p. c. canfield , and r. prozorov , phys .
b * 79 * , 134528 ( 2009 ) . j .- ph .
reid , m. a. tanatar , h. shakeripour , x. g. luo , n. doiron - leyraud , n .
ni , s. l. budko , p. c. canfield , r. prozorov , and l. taillefer , phys .
b * 82 * , 064501 ( 2010 ) .
n. doiron - leyraud , p. auban - senzier , s. ren de cotret , c. bourbonnais , d. jrome , k. bechgaard , and l. taillefer , phys .
b * 80 * , 214531 ( 2009 ) .
f. l. ning , k. ahilan , t. imai , a. s. sefat , m. a. mcguire , b. c. sales , d. mandrus , p. cheng , b. shen , and h .-
h wen , phys .
. lett . * 104 * , 037001 ( 2010 ) .
n. katayama , y. kiuchi , y. matsushita , and k. ohguchi , j. phys .
. jpn . * 78 * , 123702 ( 2009 ) .
lei fang , huiqian luo , peng cheng , zhaosheng wang , ying jia , gang mu , bing shen , i. i. mazin , lei shan , cong ren and hai - hu wen , phys.rev.b * 80 * , 140508 ( r ) ( 2009 ) .
f. rullier - albenque , d. colson , a. forget , p. thuery , s. poissonnet phys .
b * 81 * , 224503 ( 2010 ) .
s. avci , omar chmaissem , eugene a. goremychkin , stephan rosenkranz , john - paul castellan , duck - young chung , ilya s. todorov , john a. schlueter , helmut claus , mercouri g. kanatzidis , aziz daoud - aladine , dmitry khalyavin , raymond osborn arxiv:1102.1933 .
marianne rotter , michael pangerl , marcus tegel , dirk johrendt angew .
chem . int . ed . *
47 * , 7949 2008 .
kant , j. deisenhofer , a. gnther , f. schrettle , a. loidl , m. rotter , and d. johrendt , * 81 * , 014529 ( 2010 ) .
h. fukazawa , y. yamada , k. kondo , t. saito , y. kohori , k. kuga , y. matsumoto , s. nakatsuji , h.kito , p.m. shirage , k. kihou , n. takeshita , c .- h .
lee , a. iyo , and h. eisaki , j. phys .
jpn . * 78 * 083712 ( 2009 ) .
kunihiro kihou , taku saito , shigeyuki ishida , masamichi nakajima , yasuhide tomioka , hideto fukazawa , yoh kohori , toshimitsu ito , shin - ichi uchida , akira iyo , chul - ho lee , and hiroshi eisaki j. phys .
79 ( 2010 ) 124713 m. kimata , t. terashima , n. kurita , h. satsukawa , a. harada , k. kodama , a. sato , m. imai , k. kihou , c. h. lee , h. kito , h. eisaki , a. iyo , t. saito , h. fukazawa , y. kohori , h. harima , s. uji , phys .
* 105 * , 246403 ( 2010 ) . | temperature - dependent in - plane , @xmath0 , and inter - plane , @xmath1 , resistivities were measured for the iron - arsenide superconductor ( ba@xmath2k@xmath3)fe @xmath4as@xmath4 over a broad doping range from parent compound to optimal doping @xmath5 , @xmath6 .
the coupled magnetic / structural transition at @xmath7 is clearly observed for samples with @xmath826 k ( @xmath9 ) , however its effect on resistivity is much weaker than in the electron - doped ba(fe@xmath2co@xmath3)fe @xmath4as@xmath4 , and the transition leads only to a decrease of resistivity .
in addition to the feature at @xmath7 , the inter - plane resistivity shows a maximum at @xmath10200 k , which moves slightly to higher temperature with doping , revealing a trend opposite to the electron - doped materials .
a smeared feature at about the same temperature is seen in @xmath0 . for @xmath11 ,
the temperature dependence of resistivity shows systematic evolution and is close to linear at optimal doping .
this feature , being most pronounced for @xmath12 , suggests the existence of a quantum critical point close to optimal doping . |
very long fabry - perot cavities serve as measuring devices for interferometric gravitational wave detectors .
several such detectors are currently under construction @xcite .
the cavities are planned to have high circulating power and large storage time .
for example , ligo ( laser interferometer gravitational wave observatory ) fabry - perot cavities will accumulate 10 kw of power and will have roughly 1 ms of storage time . the suspended mirrors are designed to move freely along the direction of the beam propagation .
due to multi - beam interference in the fabry - perot cavity the motion of the mirrors strongly affects the light inside the cavity .
the light , in turn , affects the motion of the mirrors by exerting radiation pressure on them .
the interaction of light in the cavity with the suspended mirrors through radiation pressure gives rise to a nonlinear dynamics .
finite time of light propagation in the cavity introduces a time delay in the dynamics .
the propagation time gives rise to storage time .
thus a fabry - perot cavity is a dynamical system with delay ; such systems are known to have instabilities @xcite .
the significance of the `` spring '' action and the `` damping '' effect of the radiation pressure for the dynamics of the fabry - perot cavity was pointed out by braginskii @xcite .
the cavity with one suspended mirror driven by radiation pressure was studied experimentally by dorsel et al @xcite .
the main results reported in these papers are observations of optical bistability and mirror oscillations with frequencies determined by the radiation pressure .
these authors analyzed their results using the adiabatic approximation for the intra - cavity field . at about the same time
deruelle and tourrenc studied the fabry - perot cavity with suspended mirrors theoretically @xcite .
their analysis revealed delay - induced instability caused by the radiation pressure in the cavity .
this instability was further studied by other researchers @xcite .
stabilization of the fabry - perot cavity by a control system was discussed by meers and macdonald @xcite .
recently , the radiation pressure induced dynamics of fabry - perot cavities attracted attention of the virgo group in connection with the design of the length control system of their detector @xcite .
similar research has been done in ligo and is presented in this paper .
emerging laser gravitational wave detectors require detailed modeling and pose new questions for the study of dynamics . from a phenomenological point of view
, there is a question of what are the parameters that define the universal properties of fabry - perot cavities with suspended mirrors , and how the dynamics changes with respect to these parameters . from a point of view of applications , there is a question of how to generalize the results obtained in table - top experiments to large scale fabry - perot cavities of the gravitational wave detectors . in this paper
we attempt to provide a phenomenology of the fabry - perot cavities for modeling and optimization of the performance of ligo interferometers . due to the complexity of the subject we split the discussion into two papers . in the first paper we study various aspects of the nonlinearity in the dynamics , leaving aside the time delay instabilities . in the second paper @xcite
we consider the time delay instabilities and study the dynamics of a fabry - perot cavity with a realistic control system . in this paper
we formulate the dynamics in terms of normal coordinates : the cavity length and the cavity center of mass .
we show that a small part of the radiation pressure in the cavity , the excess force , excites the cavity center of mass . in absence of the excess radiation pressure ,
the dynamics of the cavity length is equivalent to the dynamics of a suspended mirror in a cavity , which has one mirror suspended and one mirror fixed . to study the universal properties of the cavity dynamics , such as multistability
, we introduce two parameters .
the first parameter is a control parameter which allows us to select the equilibrium state .
the second parameter characterizes strength of the radiation pressure and determines whether the system is stable or multistable and how many equilibrium states are there .
the results obtained in this paper are general and apply to any fabry - perot cavity with suspended mirrors .
numerical calculations and modeling with parameters of ligo cavities are given throughout this paper .
the paper is organized as follows . in section 2
we describe the equations of motion and define the control parameter . in section 3
we formulate the dynamics in normal coordinates . in section 4
we construct equilibrium states and introduce the specific radiation pressure . in section 5 hysteresis and instabilities are described . in section 6 the global properties of the dynamics are analyzed in terms of the effective potential . in section 7
we provide explicit formulas and numerical values for the cavity resonant frequencies . in section 8
we briefly describe nonlinear dynamics of the cavity with a control system .
the dynamical system we study in this paper is a fabry - perot cavity with two suspended mirrors , labeled by @xmath0 and @xmath1 , and a laser incident on the cavity from one side .
a suspended mirror is a rigid body with six degrees of freedom , whose dynamics depends on the suspension design . in this paper
we neglect angular degrees of freedom of the mirrors and consider only the motion of their center of mass .
we also neglect the side motion of the mirrors . in this approximation
the mirrors are always aligned to the incident beam and move parallel to themselves as simple pendula .
we specify the mirror positions by their coordinates , @xmath2 and @xmath3 , in the inertial coordinate frames as shown on figure [ coords ] .
the mirror suspension points are also moving , their coordinates are @xmath4 and @xmath5 .
the two inertial coordinate frames are separated by a large distance , the nominal cavity length , @xmath6 , which we consider fixed .
the actual length of the cavity depends on mirror positions in their respective coordinate frames : @xmath7 the large cavity length of the gravitational wave detectors makes the delay time , @xmath8 to be non - negligible .
for example , in ligo interferometers the cavities are 4 km long and the corresponding delay time is roughly 13 @xmath9s .
multibeam interference of laser in long fabry - perot cavities of gravitational wave detectors is far more complex than a similar process in short fabry - perot interferometers with fixed length which are commonly used for laser spectroscopy .
both the motion of suspended mirrors and the delay times affect the interference and give rise to a complex dynamics .
we describe dynamics of the laser in the cavity in terms of complex amplitudes of the electric field .
these complex amplitudes correspond to the traveling electro - magnetic waves and are defined at some locations .
usually the amplitudes are defined at the mirror surfaces .
however , the suspended mirrors are moving and such a choice would require account for doppler effects .
instead , we define the amplitudes at the reference planes which are fixed in the inertial coordinate frames .
our notations for the amplitudes of the electro - magnetic waves are shown in fig .
[ fields ] .
the amplitude of the wave , incident on the front mirror , is @xmath10 .
the amplitudes of the forward - propagating wave inside the cavity , defined at the two reference planes , are @xmath11 and @xmath12 .
similarly , the amplitudes of the backward - propagating wave are @xmath13 and @xmath14 .
the amplitudes of the cavity transmitted and reflected waves are @xmath15 and @xmath16 .
propagation of the laser inside the cavity is described by the delay equations : @xmath17 these and other equations for the complex amplitudes are derived in the appendix [ defe ] .
the equations for reflection and transmission of the laser at the end mirror are : @xmath18 where @xmath19 is the reflectivity and @xmath20 is the transmissivity of the end mirror . using the superposition principle we obtain similar equations for the front mirror : @xmath21 where @xmath22 and @xmath23 is the reflectivity and the transmissivity of the front mirror .
numerical values for the reflectivities and the trasmissivities of the mirrors in ligo cavities are given in the appendix [ cavpar ] .
the equations ( [ ep1]-[e4 ] ) describe dynamics of the laser interacting with the moving mirrors of the cavity .
these equations can be reduced to one equation with one field , which is called the self - consistent field .
all other fields can be expressed in terms of the self - consistent field through the field equations above .
we choose the forward - propagating field to be the self - consistent field .
let the amplitude of the self - consistent field be @xmath24 ( same as @xmath25 ) .
then the equation for dynamics of the self - consistent field is @xmath26 where @xmath27 is the relative mirror position defined as @xmath28 the quantity , @xmath29 , is the distance between the mirrors in relativistic sense .
since the speed of light is finite we can not obtain information about the instantaneous cavity length defined in eq .
( [ linst ] ) .
however , we can measure the relativistic cavity length , @xmath27 , by observation of one of the cavity fields .
this can be done using the pound - drever signal extraction scheme .
the pound - drever signal is based on observation of the cavity reflected laser @xmath30 which is defined in terms of the self - consistent field and , therefore , depends on the cavity length , @xmath27 . in the expression , eq .
( [ xdef ] ) , the delay appears in the coordinate of the end and not the front mirror .
this is because upon entering the cavity the laser first reflects off the end mirror and then reflects off the front mirror .
dynamics of the suspended mirrors is defined by the forces acting on them . since we analyze only longitudinal motion of the mirrors we consider only horizontal components of the forces .
the horizontal component of the wire tension , acting on the front mirror , is @xmath31 where @xmath32 is the mass of the front mirror and @xmath33 is the pendulum frequency .
there is a similar expression for the end mirror .
the laser circulating in the cavity and the incident laser exert pressure on the mirrors .
let the radiation pressure on the front and the end mirror be @xmath34 and @xmath35 .
these forces are defined by the amplitudes of the fields as follows : @xmath36 many control schemes , including the control system of ligo arm cavities , require actuators on the mirrors .
the actuators can be coil - magnet pairs , electro - static boards or other devices . for our analysis
the details of the actuators are not important .
we only assume that actuators are perfectly aligned and apply forces to the mirrors in the direction of the beam propagation .
let the actuator forces be @xmath37 and @xmath38 .
there is also a friction force due to energy losses in the suspension wires and the actuators .
we assume that the friction is proportional to velocity .
then the friction force acting on the front mirror can be written as @xmath39 where @xmath40 is the damping coefficient .
a similar expression can be obtained for the end mirror .
typically the damping coefficient is very small : @xmath41 .
numerical values for the parameters of ligo interferometers are given in appendix [ cavpar ] .
let us consider the sum of the pendulum restoring force and the actuator force for the front mirror : @xmath42 there are two parameters here : the position of the suspension point and the actuator force .
the mirror can be controlled through both the actuator and the suspension point .
however , the dynamics depends only on the sum of the two parameters , and not separately on each one .
therefore , we combine the two parameters into a new parameter @xmath43 this is the control parameter for the front mirror .
a similar parameter , @xmath44 , can be introduced for the end mirror .
the control parameter has dimension of length and the meaning of an equivalent displacement of the suspension point .
the equivalence is based on the conversion factor @xmath45 , which for ligo is roughly 4.2 millimeter per newton of the actuator force .
moving the suspended mirrors at high frequencies is easier using the actuator .
however , applying large displacements to the mirror at low frequencies is better through the suspension point .
therefore in a realistic control system the control parameter , @xmath46 , may obtain its high frequency component from the actuator force and its low frequency component from the position of the suspension point .
the equations of motion for the mirrors are @xmath47 these are the equations for a pair of damped harmonic oscillators driven by forces of the radiation pressure .
the oscillators are not independent : they are interacting with each other through the laser which acts as a spring connecting the mirrors .
the `` spring '' is nonlinear and has time delay .
a fabry - perot cavity has two intrinsic time scales : one is the delay time , @xmath48 , the other is the storage time @xmath49 correspondingly , there are two intrinsic frequencies defined by the cavity : the free spectral range ( @xmath50 ) and the cavity low pass frequency @xmath51 therefore , there are three dynamical regimes for the fabry - perot cavity .
let @xmath52 be a characteristic time for a dynamic process involving changes in the mirror positions and fields .
let @xmath53 be the characteristic frequencies of the process .
the three regimes are : 1 .
decoupled regime : @xmath54 ( @xmath55 ) , 2 . delay regime : @xmath56 ( @xmath57 ) , 3 .
adiabatic regime : @xmath58 ( @xmath59 ) .
the first regime corresponds to very fast changes in mirror positions and fields in the fabry - perot cavity .
the decoupled regime can be understood on the following example .
first the mirrors are at rest .
then one mirror acquires some velocity , changes its position and stops . if the entire process takes time less than @xmath48 the second mirror becomes affected by the process after the first mirror comes to rest .
therefore , during such process the mirrors are independent of each other .
this is nothing but a relativistic causality limit . in ligo interferometers
the processes with frequencies above 37.5 khz belong to the decoupled regime .
the second regime takes place when the changes in the mirror position are faster than transient processes in the cavity but not fast enough to decouple the mirrors . in this regime
the transients caused by the mirror motion never die and the cavity never reaches equilibrium . in such a regime the motion of one mirror is constantly affected by the other mirror and the delays play crucial role .
dynamic processses in ligo cavities with frequencies between 100 hz and 10 khz belong to this regime .
the third regime takes place when the mirrors move so slow that the laser transients can be neglected . in this case
we can think of the laser circulating in the cavity as being in equilibrium with the mirrors .
this is the adiabatic regime . in this regime
the delays are not important .
the adiabatic regime takes place in ligo cavities at frequencies less than 100 hz . in the adiabatic regime
the mirror motion is so slow that we can neglect the delay time in the equation for the self - consistent field eq .
( [ selfcon ] ) and obtain a solution : @xmath60 the value of the field in the adiabatic regime is completely defined by the current value of the cavity length , @xmath27 .
the power in the cavity , which we define as the momentum carried by the forward - propagating wave ( @xmath61 ) is also a function of the cavity length .
the maximum power , @xmath62 , depends on the mirror parameters and defines the cavity gain : @xmath63 the power as a function of length is described by the airy formula @xmath64 also known as airy intensity profile ( see @xcite ) .
the constant , @xmath65 , is the coefficient of finesse @xmath66 the relation between the parameter , @xmath65 , and the finesse of fabry - perot cavity is given in appendix [ airy ] .
the suspended mirrors are interacting with each other through the laser in the cavity .
therefore , the degrees of freedom which correspond to the mirrors are coupled .
however , we can formulate the dynamics in terms of new degrees of freedom which are largely independent of each other .
these new degrees of freedom are similar to the normal coordinates usually appeared in theory of coupled linear oscillators . before we formulate the dynamics in terms of the normal coordinates we reduce the number of fields entering the general expressions for the radiation pressure on the mirrors , equations ( [ ra]-[rb ] ) . the radiation pressure on the end mirror , eq .
( [ rb ] ) can be expressed entirely in terms of the self - consistent field : @xmath67 the radiation pressure on the front mirror is produced by the laser circulating in the cavity and the laser incident on the cavity , eq .
( [ ra ] ) .
a constant flux of the incident laser produces constant radiation pressure on the front mirror : @xmath68 therefore , we can write the radiation pressure on the front mirror as the sum : @xmath69 in this equation the first term implies that the same force which is pushing the end mirror now will be pushing the front mirror in the opposite direction after time @xmath48 .
the second term is proportional to the power of the incident laser and accounts for the constant radiation pressure from outside the cavity .
the last term , @xmath70 , which can be called the excess radiation pressure , accounts for the asymmetry between the mirrors and the interference between the incident laser and the laser circulating in the cavity .
the excess force can also be expressed entirely in terms of the self - consistent field and the cavity length : @xmath71 where @xmath72 , and the asterisk stands for complex conjugation . in the adiabatic regime the excess force can be written explicitly as a function of length : @xmath73 numerical estimates with ligo parameters show that the excess radiation pressure , @xmath74 , is roughly four orders of magnitude less than the radiation pressure , @xmath35 . the constant radiation pressure , eq .
( [ rin ] ) , has no effect on the dynamics , but the excess force , eq .
( [ dr ] ) , makes the radiation pressure on the mirrors unbalanced and affects the dynamics .
the normal coordinates are the position of the center of mass and the cavity length : @xmath75 note that both normal coordinates are affected by the delay due to the finite time of laser propagation in the cavity .
similarly , we introduce the control parameters for the new coordinates : @xmath76 to make the subsequent formulas simpler we hid the constant radiation pressure into the new control parameters .
also we define the reduced mass in a usual way : @xmath77 in normal coordinates the dynamical equations are the equation for the center of mass and the equation for the cavity length : @xmath78 we see that the radiation pressure mostly affects one degree of freedom : the cavity length , @xmath27 .
this is because the pressure on the front mirror is roughly equal and opposite to the pressure on the end mirror if we neglect the constant pressure from outside .
the motion of the center of mass is affected by the radiation pressure only through the excess force .
dynamics of the center of mass of free cavity is defined exclusively by the excess radiation pressure , which , in turn , is defined by the cavity length , @xmath27 . as a result ,
changes in the cavity length excite motion of the center of mass .
on the contrary , arbitrary motion of center of mass has no effect on the cavity length .
therefore we can ignore the dynamics of the center of mass and focus on the cavity length .
however , few important points must be taken into account before we abandon the motion of the center of mass .
although small motion of the center of mass ( @xmath79 m ) has no effect on the detector performance , large motion ( @xmath80 0.1 mm ) will cause variations of the actuator gains . therefore , the motion of the cavity center of mass must be suppressed to the extent defined by the actuator s range .
in particular , we must avoid exciting the motion of the center of mass by a control system . in gravitational wave detectors
the control system suppresses motion of the mirror under seismic disturbances by applying forces to the mirrors through the actuators .
therefore the control system can excite the motion of the center of mass , unless the control signals are synchronized : @xmath81 unlike the cavity length the center of mass is not observable in a strict relativistic sense . to obtain information about the cavity s center of mass
requires setting up an inertial frame , which is impossible for the ground based gravitational wave detectors .
this poses a problem of observability for control of the fabry - perot cavity , which can be stated as follows .
the cavity length is known with high precision ( better than @xmath82 m ) but the center of mass can only be measured approximately ( up to 1@xmath9 m ) .
therefore the cavity length can only be controlled approximately . since the two degrees of freedom are largely independent poor observability and control of the center of mass do not affect the dynamics of the cavity length .
therefore , as long as the motion of the center of mass remains small , it can be ignored .
the degree of freedom most important for understanding the response of fabry - perot cavity as a part of the gravitational wave detector is the cavity length @xmath27 . in the following discussion of the cavity length dynamics
we neglect the excess radiation pressure , @xmath74 . to simplify following formulas
we make the approximation : @xmath83 in the eq .
[ rb2 ] and write the radiation pressure on the end mirror as @xmath84 this is a good approximation for high reflective mirrors , such as the end mirrors of ligo cavities .
the dynamics of the cavity length is independent of the dynamics of the center of mass and can be studied separately .
the equations for the cavity length dynamics include only the cavity length and the self - consistent field : @xmath85 the iteration equation ( [ itereq ] ) allows us to calculate the amplitude @xmath24 at discrete times .
the time step is the cavity round - trip time , @xmath86 .
the value of the amplitude between the steps can be found by delaying its value at the last iteration .
these equations are equivalent to the equations of the dynamics of the fabry - perot cavity with front mirror fixed and the end mirror suspended from wires , provided the mass of the end mirror is equal to the reduced mass .
equilibrium states of the cavity are static solutions of the equations ( [ pendeq ] ) and ( [ itereq ] ) . for static states these equations simplify and can be written as @xmath87 the complex field , @xmath88 , can be eliminated from the equations ( [ equilx ] ) and ( [ equile ] ) and the equilibrium condition can be written entirely in terms of the cavity length : @xmath89 here @xmath90 is the maximum radiation pressure on the end mirror which corresponds to the maximum power in the cavity : @xmath91 the equation ( [ equilx ] ) shows that in equilibrium the radiation pressure is balanced by the pendulum restoring force .
numerical values of the power and the radiation pressure in ligo fabry - perot cavities are given in the table [ mags ] .
.power and radiation pressure in ligo fabry - perot cavities [ cols="<,^,^,^",options="header " , ]
we approximate the laser by a plane electro - magnetic wave with electric field , @xmath92 , which is a solution of the 1-dimensional wave equation : @xmath93 let @xmath94 be the electric field corresponding to the right - moving wave in the cavity and @xmath95 be the electric field corresponding to the left - moving wave .
we define the complex amplitudes at the reference planes , located at @xmath96 and @xmath97 .
the amplitudes of the right - moving wave , @xmath11 and @xmath12 , are defined by the following equations : @xmath98 similarly , we define the amplitudes,@xmath13 and @xmath14 , for the left - moving wave : @xmath99 since the wave functions satisfy the wave equation their values at different locations can be obtained by delay : @xmath100 therefore , the amplitudes at the two reference planes are related to each other : @xmath101 if we assume that @xmath102 where @xmath103 is an integer , we obtain the equations ( [ ep1]-[ep2 ] )
. the equation for the reflection off the end mirror can be obtained from the condition of continuity of the wave function at the mirror surface : @xmath104 therefore , at the reference plane @xmath97 : @xmath105 this equation can now be written in terms of the amplitudes : @xmath106 if we neglect the small time delay due to the mirror motion , @xmath107 , we obtain the equation ( [ e2 ] ) .
other equations can be derived similarly .
here we present few basic properties airy function , which are used throughout this paper .
the airy function , @xmath108 is a postitive , periodic function of @xmath109 with the period @xmath110 .
the airy function depends on one parameter , @xmath65 , a positive number called the coefficient of finesse .
[ airyfs ] shows airy functions with different coefficients of finesse .
the half - width , @xmath111 , is the phase , at which the airy function equals to half of its maximum value @xmath112 the full width is equal to @xmath113 .
finesse is the ratio of the separation between the peaks to the full width of the peak @xmath114 a slope of a tangent line to airy function at point @xmath109 is defined by the derivative of airy function @xmath115 the inflection points of airy function can be found by solving the equation : @xmath116 .
they are located at @xmath117 , where @xmath118 the greatest slope the tangent line achieves at the left inflection point @xmath119 .
this maximum slope is @xmath120^{1/2}}{\left ( \frac{1}{4}\sqrt{9f^2 + 4f + 4 } - \frac{3}{4}f - \frac{3}{2}\right)^2}. % \ ] ] the expression for the inflection points , eq .
( [ infl ] ) , and the maximum slope , eq .
( [ maxda ] ) , are quite complicated , therefore , it useful to find simple approximate formulas for these quantities . for high finesse cavities ( large @xmath65- limit )
such approximations can be easily found : @xmath121 any straight line , which has a slope greater than the maximum slope , intersects the airy function at only one point .
if the slope of the line is less than the maximum slope the line can intersect the airy function at more than one point . | the dynamics of fabry - perot cavity with suspended mirrors is described .
the suspended mirrors are nonlinear oscillators interacting with each other through the laser circulating in the cavity .
the degrees of freedom decouple in normal coordinates , which are the position of the center of mass and the length of the cavity .
we introduce two parameters and study how the dynamics changes with respect to these parameters .
the first parameter specifies how strong the radiation pressure is .
it determines whether the cavity is multistable or not .
the second parameter is the control parameter , which determines location of the cavity equilibrium states .
the equilibrium state shows hysteresis if the control parameter varies within a wide range .
we analyze stability of the equilibrium states and identify the instability region .
the instability is explained in terms of the effective potential : the stable states correspond to local minima of the effective potential and unstable states correspond to local maxima .
the minima of the effective potential defines the resonant frequencies for the oscillations of the cavity length .
we find the frequencies , and analyze how to tune them .
multistability of the cavity with a feedback control system is analyzed in terms of the servo potential .
the results obtained in this paper are general and apply to all fabry - perot cavities with suspended mirrors . |
it is widely believed that the new particle discovered at fermilab in 1995 @xcite is the long sought top quark .
its behavior is certainly consistent with standard model ( sm ) expectations for top , but many of its properties are still only poorly known . in particular , the top quark s electric charge one of the most fundamental quantities characterizing a particle has not been measured yet .
it still remains not only to confirm that the discovered quark has charge @xmath7 ( and hence the expected sm quantum numbers ) , but also to measure the strength of its electromagnetic ( em ) coupling to rule out anomalous contributions to its em interactions .
indeed , one alternative interpretation has not yet been ruled out : that the new particle is a charge @xmath5 quark . in
the published top quark analyses of the cdf and d collaborations ( see ref .
@xcite for a review ) , the correlations of the @xmath3-quarks and the @xmath8 bosons in @xmath9 are not determined . as a result
, there is a twofold ambiguity in the pairing of @xmath8 bosons and @xmath3-quarks , and , consequently , in the electric charge assignment of the `` top quark '' .
besides the standard model ( sm ) assignment , @xmath10 , @xmath11 is also conceivable , in which case the `` top quark '' would actually be an exotic quark with charge @xmath12 .
interpreting the particle found at fermilab as a charge @xmath5 quark is consistent with current precision electroweak data .
current @xmath13 and @xmath14 data can be fit with a top quark of mass @xmath15 gev , provided that the right - handed @xmath3-quark mixes with the isospin @xmath16 component of an exotic doublet of charge @xmath17 and @xmath5 quarks , @xmath18 @xcite .
if the top quark would have a mass of @xmath15 gev , it would have escaped detection in run i of the tevatron . in this scenario ,
the particle discovered in run i is the @xmath19 .
a direct measurement of the top quark charge would provide a definitive test of this interpretation .
there are several techniques to determine the electric charge of the top quark , @xmath20 , in future collider experiments . at a linear @xmath21 collider , the full set of form factors describing the most general @xmath22 vertex function compatible with lorentz invariance
can be probed in @xmath23 @xcite ( similarly for a muon collider ) .
a photon - photon collider would provide even more precision via the two electromagnetic vertices in @xmath24 .
the status of future lepton and photon colliders is unfortunately still somewhat uncertain , but in the meantime top quarks will be produced copiously at the fermilab tevatron and the cern lhc . at these hadron colliders ,
@xmath25 production is so dominated by the qcd processes @xmath26 and @xmath27 that a measurement of the @xmath22 form factors via @xmath28 is hopeless .
instead , one can measure the top charge by measuring the charges of its decay products , the final state @xmath3-jets and @xmath8 bosons .
one can further attempt to determine not only the top quark s charge but also the em coupling strength by directly studying top s electromagnetic interactions through photon radiation in @xmath0 events . the first method measuring the top charge by reconstructing the charges of its decay products
is difficult to realize for @xmath29 all jet decays ; however for di - lepton decays , @xmath30 , and for semileptonic decays , @xmath31 , it should be feasible .
the @xmath3-jet charge can be determined from a measurement of the charges associated with the tracks in the jet . a preliminary measurement of the @xmath3-jet charge using run i cdf data @xcite shows a slight preference for the sm top charge assignment .
the direct measurement of the @xmath3-jet charge has the disadvantage that many tagged @xmath3-quarks are needed to obtain a statistically significant result . in a given event , missing or out - of - cone tracks can distort the measurement .
information on the @xmath3-jet charge can also be obtained from the charge - sign of the lepton , @xmath32 , in events where the @xmath3-quark is identified via a soft lepton tag , i.e. where the @xmath3-quark decays semileptonically , @xmath33 . in the absence of @xmath34 and @xmath35 mixing , the charge - sign of the lepton is directly correlated with the charge - sign of the parent @xmath3-quark .
the difficulties associated with using soft lepton tagged @xmath3-quarks are low efficiency due to the small @xmath36 branching ratio and the presence of wrong sign leptons originating from @xmath34 and @xmath35 mixing , which one has to correct for . in addition , if only the charge of the @xmath3 decay lepton is measured , only the sign but not the magnitude of the charge of the @xmath3-quark is determined . in this paper
, we explore the possibility of measuring the electric charge of the top quark at hadron colliders through photon radiation in @xmath0 events , assuming that the em coupling strength takes its sm value events . ] . because top quarks can radiate photons in both top quark production and top decay , we consider both processes : @xmath37 we concentrate on the lepton+jets mode , are cleaner than lepton+jets events , their branching fraction is about a factor 6 smaller than that of the lepton+jets mode .
the all - jets mode has a larger branching ration than lepton+jets , but is plagued by a large qcd background and , therefore , is also not considered here . ] , @xmath38 @xmath39 , and assume that both @xmath3-quarks are tagged .
we present results for the sm top quark and for comparison we also consider a charge @xmath5 quark of the type discussed in @xcite .
photon radiation in top quark events was not observed by the tevatron experiments in run i. in run ii , an integrated luminosity of @xmath40 is envisioned , and a sufficient number of @xmath41 events may be available to measure the top quark charge . at the lhc , where the cross section for @xmath0 production is more than a factor 100 larger than at the tevatron , one can hope for a determination of @xmath20 with a precision of about 10% .
details of our calculation are given in sec .
ii . in sec .
iii we present results of our numerical studies .
in addition to total cross sections , distributions for the transverse momentum of the photon and the separation between the photon and the @xmath3-quarks in the pseudorapidity azimuthal angle plane are found to be sensitive to @xmath20 . in sec . iv we perform a quantitative analysis of
how well the charge of the top quark can be measured at the tevatron and the lhc . finally , in sec .
v , we present our conclusions .
our calculation is carried out at the tree level .
we assume that both @xmath3-quarks are tagged with a combined efficiency of 40% , which is included in all our numerical results .
top quark and @xmath8 boson decays are treated in the narrow width approximation . in this approximation
, there are three contributions to @xmath42 : * radiation in top production : @xmath43 . * radiative top decay : @xmath44 . here either the @xmath45 or @xmath46 quark decays radiatively .
we denote these decays generically by @xmath47 .
* radiative @xmath8 decay : @xmath48 . in this case ,
one of the @xmath8 bosons decays radiatively , @xmath49 or @xmath50 .
both @xmath1 production and radiative top decays are sensitive to @xmath20 . on the other hand ,
radiative @xmath8 decays are not , and should be suppressed by cuts .
this can be achieved by requiring that @xmath51 where @xmath52 denotes the missing transverse momentum originating from the neutrino in the event which is not observed , and @xmath53 is the invariant mass of the @xmath54 system .
@xmath55 is the @xmath56 cluster transverse mass , which is given by @xmath57 where @xmath58 and @xmath59 are the transverse momentum and the invariant mass of the @xmath60 system , respectively . for @xmath49 , the cluster transverse mass sharply peaks at the @xmath8 boson mass , @xmath61 . in the narrow width approximation , the cuts listed in eq .
( [ eq : radw ] ) completely eliminate the contributions from radiative @xmath8 decays to the @xmath41 final state .
we thus ignore radiative @xmath8 decays in the following .
the matrix elements for @xmath62 are calculated using madgraph @xcite and the helas library @xcite , including the spin correlations for the subsequent top decays . to compute the matrix elements for @xmath63 and its charge conjugate sibling , we use the crossed form of the @xmath64 matrix elements of ref . @xcite .
we ignore spin correlations in the decay of the @xmath8 boson in @xmath65 . for the @xmath66 decay ,
spin correlations are taken into account .
all numerical results presented below are obtained summing over electron and muon final states .
the top quark charge , @xmath20 , is treated as a free parameter in our calculation .
the electric charge of the @xmath3-quark , @xmath67 , is related to @xmath20 by @xmath68 where @xmath69 is the charge of the @xmath8 boson .
the sign of @xmath70 depends on whether the third component of the weak isospin of the `` top quark '' is @xmath71 , as in the sm , or @xmath72 as in the scenario of ref .
we assume the @xmath73 coupling to be of @xmath74 form and br@xmath75 .
the fermions originating from @xmath8 decays are assumed to be massless .
we note for future reference that the differential cross sections for @xmath76 and for @xmath77 with one top quark decaying via @xmath78 are invariant under the transformation @xmath79 since interference effects between the incoming light quarks and the final state top quarks in @xmath80 are very small ( they contribute less than 1% to the cross section ) , the cross section for @xmath81 is to a very good approximation invariant under the transformation ( [ eq : trans ] ) .
this will be relevant when we estimate how well @xmath20 can be measured in future tevatron and lhc experiments ( see sec .
iv ) . for all our numerical simulations
we have chosen the set of sm input parameters to be : @xmath82 gev @xcite , @xmath83 gev , @xmath84 gev , @xmath85 and @xmath86 @xcite .
we employ mrsr2 parton distribution functions @xcite , selecting the value of the factorization scale to be @xmath87 , where @xmath88 is the squared parton center of mass energy .
calculations are carried out for @xmath89 collisions at @xmath90 tev ( tevatron ) and @xmath91 collisions at @xmath92 tev ( lhc ) .
to simulate detector response , we impose the following common transverse momentum , rapidity and separation cuts at the tevatron and lhc : @xmath93 where @xmath94 . @xmath95 and
@xmath96 denote the two jets originating from the hadronically decaying @xmath8 in the event .
as mentioned above , we assume that both @xmath3 quarks are tagged ( 40% combined efficiency ) so that @xmath3 and @xmath97 are distinguishable from @xmath95 and @xmath96 . in addition , we require that @xmath98 here , @xmath99 is the pseudo - rapidity , and @xmath100^{1/2}\ ] ] is the separation between two particles @xmath101 and @xmath102 in the pseudorapidity azimuthal angle plane .
the transverse momentum cuts listed in eqs .
( [ eq : top31 ] )
( [ eq : top35 ] ) and in eq .
( [ eq : top5 ] ) for the lhc are sufficient for running at low luminosity , @xmath103 .
for operation at the design luminosity of @xmath104 , the @xmath105 cuts must be increased due to pile - up effects caused by the large number of interactions per beam crossing . in the following we only consider @xmath41 production at the lhc in the low luminosity environment .
we do not consider any potential background processes here .
the dominant background is expected to originate from @xmath106 jets production and should be manageable in much the same way as the @xmath107 jets background for @xmath108 production .
in this section we present our results for total cross sections and various photon distributions .
we look at contributions from individual subprocesses ( radiative production , radiative decays ) and compare results for the sm top quark to those for an exotic charge @xmath5 top ( as in ref .
@xcite ) in order to determine which quantities are sensitive to the value of @xmath20 .
the events passing the cuts listed in eqs .
( [ eq : radw ] ) and ( [ eq : top31 ] ) ( [ eq : top5 ] ) can be split into three separate samples , each designed to enhance one of the three subprocesses : ( i ) radiation in top production ( `` @xmath1 selection cuts '' ) , ( ii ) radiative top decay with leptonically decaying @xmath8 ( `` @xmath109 selection cuts '' ) , and ( iii ) radiative top decay with hadronically decaying @xmath8 ( `` @xmath110 selection cuts '' ) .
the event sample for radiation in top production ( @xmath1 selection cuts ) is obtained by suppressing radiative top decay events .
we do this by selecting events which satisfy @xmath111 here , @xmath112 and @xmath113 .
we note that by imposing the cuts on either @xmath3-quark combination , we conservatively assume that the charges of the individual @xmath3 quarks have not been determined . the total cross sections for @xmath114 and @xmath115 , imposing the @xmath1 selection cuts of eq . ( [ eq : top51 ] ) and the cuts described in sec .
ii , are listed in table [ tab : one]a for the tevatron and the lhc . at the tevatron ,
the @xmath116 cross section is completely dominated by @xmath117 annihilation . as a result , photon radiation off the initial state quarks constitutes an irreducible background which limits the sensitivity of the cross section to @xmath20 . in contrast , at the lhc more than 75% of the @xmath1 cross section originates from gluon fusion .
since gluons do not radiate photons , the lhc cross section scales approximately with @xmath118 .
radiative top decays contribute about 10 20% ( 1% ) to the cross section for @xmath114 ( @xmath115 ) in this phase space region . at both
the tevatron and lhc , increasing @xmath119 increases the total cross section with @xmath1 selection cuts .
the photon transverse momentum distributions for the individual contributions to the @xmath41 differential cross section at the tevatron ( lhc ) for @xmath1 selection cuts are shown in fig .
[ fig : fig0 ] ( fig . [ fig : fig0a ] ) for both @xmath114 and @xmath115 , keeping the same vertical scale for both charges .
the cross section from @xmath0 production followed by @xmath109 is about 20% larger that from @xmath120 followed by @xmath121 over most of the @xmath105 range considered . at the lhc ,
the @xmath122 distribution for the radiative top decay channels drops much faster than that for @xmath123 .
a sample for radiative top decay with leptonically decaying @xmath8 ( @xmath109 selection cuts ) is obtained by requiring that @xmath124 in addition to ( [ eq : top51 ] ) being _ not _ satisfied .
these cuts enhance the contribution of the process @xmath125 , with @xmath109 . similarly , to get events with radiative top decay and hadronically
decaying @xmath8 ( @xmath121 selection cuts ) we impose @xmath126 and require that ( [ eq : top51 ] ) is not fulfilled . we thereby obtain an event sample in which the process @xmath125 , with @xmath127 is enhanced .
the cuts are imposed on either of the @xmath3-quark combinations , i.e. we assume that it has not been determined whether the @xmath3 or the @xmath97 originates from the radiative top quark decay . since the finite experimental resolution of detectors significantly broadens the breit - wigner resonance for @xmath127 , we have chosen a relatively large window for @xmath128 around @xmath129 . requiring that events do not satisfy ( [ eq : top51 ] )
significantly improves the efficiency of the @xmath78 selection cuts .
the cross sections for @xmath130 imposing @xmath109 ( eq . ( [ eq : top6 ] ) ) and @xmath127 selection cuts ( eq . ( [ eq : top7 ] ) ) are shown in table [ tab : one]b and [ tab : one]c for @xmath114 and @xmath115 .
the total cross sections are smaller than the corresponding numbers for @xmath1 selection cuts
. moreover , the cross section for @xmath0 production followed by radiative @xmath45 decay _ decreases _ by a factor 3 to 4 when the top quark charge is varied from @xmath114 to @xmath115 .
this is due to interference effects between the diagrams where the photon is radiated from the @xmath45-quark line and the diagrams where the photon is either emitted from the final state @xmath8 or the @xmath3-quark . for the photon @xmath105 cut
used , a significant fraction of the events in the sm at the tevatron in the regions defined by the @xmath109 and @xmath127 selection cuts originates from @xmath1 production . for @xmath115 ,
@xmath1 production is in fact the dominant source in these regions .
nonetheless , the net result is a decrease in the total cross section when we switch from @xmath114 to @xmath115 .
recall that , in contrast , the cross section increased with the change in top charge for @xmath1 selection cuts .
this suggests that the _ ratio _ of the cross section associated with radiation in top production to the radiative decay cross section would be very sensitive to the top quark charge .
this issue is explored in the next section . combining the cross sections for all three phase space regions
, we expect about 60 @xmath41 events at the tevatron for 20 fb@xmath4 .
the number of events expected at the lhc is about an order of magnitude larger : 550 @xmath41 events for 10 fb@xmath4 .
returning to the radiative decay sample , we show in fig .
[ fig : one ] the individual contributions to the @xmath41 differential cross section for @xmath109 selection cuts at the tevatron for @xmath114 and @xmath115 as a function of the photon @xmath105 .
the individual contributions at the lhc are shown in fig .
[ fig : two ] .
we see that the contribution from @xmath1 production in the sm at the tevatron exceeds that from @xmath120 with @xmath131 only for @xmath132 gev ( fig .
[ fig : one]a ) .
increasing the photon @xmath105 cut thus improves the efficiency of the @xmath131 selection cuts , at the cost of a reduced event rate . because the cross sections for the radiative top decay processes decrease whereas the @xmath1 rate increases when @xmath115 , @xmath1 production is the largest contribution to the differential cross section over the entire @xmath122 range considered at both the tevatron and the lhc ( figs .
[ fig : one]b and [ fig : two]b .
the differential cross section of the @xmath133 `` background '' drops much faster with @xmath122 than those of the two other contributions for large photon transverse momenta .
when @xmath127 selection cuts are imposed , qualitatively similar results are obtained , with the roles of the @xmath109 and @xmath127 curves exchanged and somewhat reduced contributions from @xmath1 production .
the preceding results are summarized in figs .
[ fig : three ] and [ fig : four ] . fig .
[ fig : three]a shows the total sm ( @xmath134 ) cross section for @xmath135 at the tevatron , with all contributions from radiation in top production and radiative decays combined .
the three curves correspond to the three sets of selection cuts .
[ fig : four]a shows the corresponding results for the lhc ( with @xmath91 initial state ) .
we see in these figures , as we have already seen in table [ tab : one ] , that the @xmath1 selection cuts ( solid curves ) result in higher cross sections at both colliders than the cuts for radiative decays @xmath78 . in figs .
[ fig : three]a and [ fig : four]a we also see that the photon distribution imposing @xmath109 and @xmath127 selection cuts drops much faster than the @xmath122 distribution in the phase space region defined by the @xmath1 selection cuts .
this is a consequence of the phase space restriction on the photon energy in radiative top quark decays . in figs .
[ fig : three]b and [ fig : four]b we compare cross sections for @xmath134 ( solid curves ) and @xmath115 ( dashed curves ) for the tevatron and lhc , respectively . in each figure
the upper pair of curves corresponds to @xmath1 selection cuts and the lower pair includes events from both radiative decay regions combined .
the figures show that at both machines , the effect of increasing @xmath119 is to increase the cross section in the @xmath1 cuts region and to decrease it in the radiative decay regions , for all values of photon @xmath105 . at the tevatron , however , since the contributions from radiative top quark decays are small if @xmath1 selection cuts are imposed , changing the top quark charge has little influence on the shape of the photon @xmath105 distribution in this region ( upper curves in fig .
[ fig : three]b ) . on the other hand , at small values of @xmath122 ,
@xmath1 production may contribute significantly to the cross section in the regions defined by the @xmath136 and @xmath127 selection cuts ( see table [ tab : one ] ) .
the shape of the photon transverse momentum distribution in these regions thus may be sensitive to @xmath20 ( lower curves in fig .
[ fig : three]b ) . the curves at the lhc ( fig .
[ fig : four]b ) also exhibit some shape dependence , albeit less pronounced than at the tevatron .
we can look for sensitivity to @xmath20 in other distributions as well . in @xmath78 ,
emission of the photon from the @xmath3-quark line leads to a collinear enhancement at small values of the separation between the photon and the @xmath3-quark , @xmath137 , which depends on @xmath138 .
therefore , in radiative top decays , the shape of the @xmath137 distribution is expected to be sensitive to the charge of the top quark .
in contrast , the photon and the @xmath3-quark originating from the @xmath139 decay in @xmath140 events are essentially uncorrelated . in @xmath1 events , the photon and _ both _
@xmath3-quarks are largely uncorrelated . unfortunately , on an event by event basis
, it is difficult to determine whether the @xmath3 or the @xmath97 quark originates from the radiative top decay .
however , due to the collinear enhancement for small values of @xmath141 in radiative top decays , @xmath142\ ] ] should coincide with the photon
@xmath3-quark separation in radiative top decays for small values of @xmath141 .
alternatively , the @xmath143 distribution , where @xmath144 is the @xmath3 or @xmath97-quark with the smaller transverse momentum @xmath145\ ] ] may be considered . since a @xmath3-quark which radiates a photon looses energy , the @xmath144 in @xmath140 frequently originates from the radiative top decay .
it turns out that the @xmath143 distribution more closely resembles the true @xmath137 distribution .
figure [ fig : five ] shows a comparison of the @xmath146 and @xmath143 distributions at the tevatron in the sm case , imposing @xmath147 selection cuts . in each case , the individual distributions of the three contributing subprocesses are shown .
qualitatively similar results are obtained for the lhc case , and for @xmath133 selection cuts . due to the collinear enhancement at small opening angles between the @xmath3-quark and the photon , the @xmath137 distribution rises very quickly with decreasing values of @xmath137 for the radiative top decay processes . for @xmath1 production
no collinear enhancement occurs for either the @xmath148 or the @xmath149 distribution . in fig .
[ fig : five]a , we have chosen the @xmath150 distribution for @xmath1 production ( dotted line ) .
very similar results are obtained if the @xmath148 distribution is chosen instead . except for the steepness of the distribution at small @xmath143 for @xmath109 ,
the shapes of the @xmath137 and @xmath143 distributions are found to agree reasonably well . in fig .
[ fig : five ] we have increased the minimum photon transverse momentum requirement of eq .
( [ eq : top4 ] ) to @xmath151 gev . as we have mentioned before
, @xmath1 production contributes significantly to the cross section at the tevatron in the region defined by the @xmath152 selection cuts for the cuts specified in sec .
the @xmath1 contribution can be reduced by increasing the @xmath105 cut on the photon ( see fig . [
fig : one ] ) , at the cost of a decreased number of events .
increasing the photon transverse momentum cut to @xmath151 gev reduces the @xmath1 cross section for @xmath153 selection cuts by about a factor 2 , whereas the contributions involving radiative top decays decrease by only 30% .
we compare the @xmath141 distributions for the two top quark charges in figures [ fig : six ] and [ fig : seven ] .
we display the normalized distribution @xmath154 for @xmath155 with @xmath114 and @xmath115 at the tevatron and lhc , respectively .
normalizing the distributions to unit area emphasizes the dependence of the shape of the @xmath143 distribution to the top quark charge . for a charge @xmath5 quark ,
the @xmath143 distribution rises relatively faster in the region @xmath156 , but is flatter for @xmath157 , than if @xmath114 . at small values of @xmath143 ,
emission from the @xmath3-quark line dominates .
since the _ magnitude _ of @xmath67 is the same for @xmath114 and @xmath115 , the differential cross section is similar in both cases .
however , the total cross section for @xmath115 within cuts is significantly smaller than for a sm top quark charge assignment ( see table [ tab : one ] ) , due to destructive interference between the contributing feynman diagrams .
this leads to a larger normalized differential cross section for @xmath115 at small @xmath143 . in the region @xmath157 ,
the destructive interference is responsible for the flatness of the @xmath154 distribution in the @xmath115 case . for @xmath158 , the @xmath3-quark with the smaller @xmath105 most of the time coincides with the bottom quark from the ( regular ) @xmath139 decay and ,
thus , carries little information on @xmath20 .
the shape of the @xmath143 distribution in the phase space region defined by the @xmath127 selection cuts ( figs .
[ fig : six]b and [ fig : seven]b ) is found to be somewhat more sensitive to @xmath20 than in the case when @xmath109 selection cuts are imposed ( figs .
[ fig : six]a and [ fig : seven]a ) . at the tevatron ( fig .
[ fig : six ] ) , we have again imposed a @xmath151 gev cut .
the shape of the @xmath143 distribution for @xmath140 at the lhc is in general more sensitive to the top quark charge than at the tevatron .
in this section we determine quantitative bounds on @xmath159 which one expects to achieve with 20 fb@xmath4 of data at the tevatron in run ii , and 10 fb@xmath4 at the lhc , analyzing the photon @xmath105 and the @xmath143 distributions .
we also briefly consider the ratio of the cross section associated with radiation in top production to the radiative decay cross section which may be useful for small data sets .
in our analysis of differential cross sections , we calculate 95% confidence level ( c.l . )
limits performing a @xmath160 test .
the statistical significance is calculated by splitting the selected distribution into a number of bins , each with typically more than five events . in each bin
the poisson statistics are approximated by a gaussian distribution .
we use the photon transverse momentum distribution in the phase space region defined by the @xmath1 selection cuts of ( [ eq : top51 ] ) , and the @xmath122 and @xmath161 distributions of the combined radiative top decay regions , defined by requiring that events satisfy ( [ eq : top6 ] ) or ( [ eq : top7 ] ) but not ( [ eq : top51 ] ) . we impose the cuts described in sec .
ii , except for the @xmath161 distribution , where we replace the @xmath105 cut on the photon of eq . (
[ eq : top4 ] ) with @xmath151 gev .
we combine channels with electrons and muons ( both charge combinations ) in the final state .
we also make use of the approximate symmetry of the cross sections under the transformation ( [ eq : trans ] ) and define the @xmath3-quark charge to be @xmath162 .
confidence levels calculated for @xmath163 will thus automatically apply for a charge @xmath5 quark which decays into a @xmath3-quark and a @xmath164 .
our calculation does not include qcd corrections which are expected to modify the cross section by @xmath165 . in order to derive realistic limits ,
we therefore allow for a normalization uncertainty of @xmath166 of the sm cross section .
the expression for @xmath160 which is then used to compute confidence levels is given by @xcite @xmath167 where @xmath168 is the number of bins , @xmath169 is the number of events for a given @xmath170 , and @xmath171 is the number of events in the sm in the @xmath101th bin .
@xmath172 reflects the uncertainty in the normalization of the sm cross section within the allowed range , and is determined by minimizing @xmath160 : @xmath173 with @xmath174 in table [ tab : two ] we display the resulting 95% confidence level limits for @xmath170 expected from the tevatron run ii for an integrated luminosity of 20 fb@xmath4 , and from the lhc for 10 fb@xmath4 .
the last row shows the combined limits from the three individual distributions studied here .
possible backgrounds are assumed to be subtracted . at both the tevatron and the lhc , the @xmath122 and @xmath161 distributions in the combined radiative top decay regions
yield rather similar limits for @xmath170 .
as mentioned before , @xmath1 production at the tevatron is completely dominated by @xmath117 annihilation and photon radiation from the initial state quarks limits the sensitivity of the cross section to @xmath20 .
therefore , the bounds which can be achieved at the tevatron from analyzing the photon @xmath105 distribution in the phase space region defined by the @xmath1 selection cuts are significantly weaker than those which can be obtained from either the @xmath122 or the @xmath161 distribution in the combined radiative top decay regions .
distributions more than those for the photon @xmath105 . ] at the lhc , most of the @xmath1 cross section originates from gluon fusion , and the limits from the photon @xmath105 distribution for @xmath116 selection cuts are similar to those which can be achieved in the phase space regions where radiative top decays dominate .
while a precise measurement of the top quark charge will not be possible in run ii of the tevatron , even with an integrated luminosity of 20 fb@xmath4 , we note that a charge @xmath5 top quark can be ruled out at @xmath6 c.l . when the limits from the different distributions and phase space regions are combined . at the lhc , with 10 fb@xmath4
, a measurement of the top quark charge with an accuracy of about 10% appears to be feasible . as we pointed out in sec .
ii , the @xmath105 cuts we use are suitable for running at low luminosity at the lhc , i.e. for @xmath175 . at high luminosity , @xmath176 , the transverse momentum cuts must be increased due to pile - up effects caused by the large number of interactions per beam crossing .
the @xmath41 cross section is quite sensitive to these cuts .
for example , increasing the @xmath105 cut of the @xmath3-quarks to 30 gev and the photon , jet and missing @xmath105 cuts to 50 gev reduces the @xmath1 and @xmath140 cross sections by a factor @xmath177 and @xmath178 , respectively .
it is thus not clear whether running at high luminosity will improve the measurement of the top quark electric charge at the lhc .
the bounds on @xmath170 are in general quite sensitive to @xmath179 .
if the normalization uncertainty can be reduced to @xmath180 , the limits listed in table [ tab : two ] can be improved by up to a factor 2 .
further improvements may result from using more powerful statistical tools than the simple @xmath160 test we performed . for integrated luminosities smaller than 20 fb@xmath4 at the tevatron
, the limited number of events makes a detailed analysis of differential cross sections difficult . in this case
, the ratio of the cross section associated with radiation in top production to the radiative decay cross section , @xmath181 , may be more useful .
many experimental and theoretical uncertainties , for example , those associated with the particle detection efficiencies , the uncertainty in the integrated luminosity , or the uncertainties from the choice of scale or of the parton distribution functions , are expected to cancel , at least partially , in @xmath181 .
the cross section ratio is quite sensitive to @xmath20 . from the individual cross sections listed in table [ tab : one ] one finds @xmath182 for @xmath114 , and @xmath183 for @xmath115 at the tevatron . for an integrated luminosity of 10 fb@xmath4 ,
the expected statistical uncertainty at the tevatron is @xmath184 and a charge @xmath5 top quark can thus be ruled out at @xmath185 c.l . using the cross section ratio , if one ignores possible systematic uncertainties . at the lhc ,
the value of @xmath181 is even more sensitive to @xmath20 than at the tevatron , but enough events are expected to allow a more detailed analysis .
like most of its fundamental quantum numbers , the electric charge of the top quark has not been measured so far .
alternative interpretations for the particle we believe is the charge @xmath7 isospin partner of the @xmath3-quark are thus not ruled out .
for example , since the correlations of the @xmath3-quarks and the @xmath8-bosons in @xmath186 are not determined , it is conceivable that the `` @xmath45-quark '' observed at the tevatron is an exotic quark , @xmath19 , with charge @xmath5 which decays via @xmath187 .
this interpretation is consistent with current precision electroweak data @xcite . in order to determine the charge of the top quark
, one can either measure the charge of the @xmath3-jet , or investigate photon radiation in @xmath0 events .
the latter method actually measures a combination of the em coupling strength and the charge quantum number .
combining the results of the two methods will thus make it possible to determine both quantities . in this paper
we have explored the possibility of measuring @xmath20 in @xmath188 , assuming that both @xmath3-quarks are tagged .
our analysis makes use of both @xmath1 production and radiative top quark decays , and is carried out at the tree level .
the em coupling strength is assumed to take its sm value .
top quark and @xmath8 boson decays are treated in the narrow width approximation .
contributions from radiative @xmath8 decays can be suppressed by simple phase space cuts ( see eq .
( [ eq : radw ] ) ) .
the remaining event sample can be separated into a @xmath1 sample , a @xmath25 , @xmath189 sample , and a @xmath25 , @xmath190 sample by imposing invariant mass and cluster transverse mass cuts ( see eqs .
( [ eq : top51 ] ) ( [ eq : top7 ] ) ) . in the phase space region dominated by @xmath1 production , the cross section increases if the magnitude of @xmath20 increases . at the tevatron , @xmath117 annihilation dominates and photon radiation off the incoming quarks constitutes an irreducible background which limits the sensitivity to @xmath20 .
in contrast , at the lhc , gluon fusion dominates , and the @xmath1 cross section scales approximately with @xmath118 . in the phase space regions which enhance @xmath0 production with one of the top quarks decaying radiatively ( @xmath78 ) , interference effects between the feynman diagrams where
the photon is emitted from the top quark , the @xmath8 boson and the @xmath3-quark line may cause the cross section to either decrease or increase with @xmath191 .
for example , for @xmath115 , the @xmath192 cross section via radiative top quark decays is reduced by a factor of @xmath193 to @xmath194 .
the changes in the individual cross sections are reflected in the photon transverse momentum distribution in the various phase space regions .
we also found that the @xmath143 distribution , where @xmath144 is the @xmath3 or @xmath97-quark with the smaller @xmath105 , is sensitive to the charge of the top quark in the phase space regions dominated by @xmath0 production where one of the top quarks decays radiatively . in order to determine
how well one can hope to measure the top quark charge using photon radiation in top quark events in future tevatron and lhc experiments , we have performed @xmath160 tests of the @xmath122 and the @xmath143 distributions in the various phase space regions of interest . at the tevatron , with an integrated luminosity of 20 fb@xmath4
, one will be able to exclude at @xmath6 c.l .
the possibility that an exotic quark @xmath19 with charge @xmath5 and not the sm top quark has been found in run i. for smaller integrated luminosities , the number of events expected is very small and it will be difficult to perform a quantitative analysis using differential cross sections . in this case , the ratio of the cross section associated with radiation in top production to the radiative decay cross section may be a useful tool . at the lhc with 10 fb@xmath4 obtained at @xmath175 , it should be possible to measure the electric charge of the top quark with an accuracy of about 10% . for comparison , at a linear collider with @xmath195 gev and @xmath196 fb@xmath4 ,
one expects that @xmath20 can also be measured with a precision of about 10% @xcite .
finally , at a @xmath197 collider it is conceivable that the top quark charge can be determined with an accuracy of better than 1% , if the @xmath24 cross section can be measured with a precision of 2% @xcite .
the measurement of the electric charge of the top quark at the tevatron using photon radiation in top quark events is severely limited by statistics .
even with an integrated luminosity of 20 fb@xmath4 , @xmath20 can only be determined with a precision of @xmath198 . however , as we pointed out in sec .
i , information on the electric charge of the top quark may also be obtained from a measurement of the charge of the @xmath3-jets , and the charge sign of the leptons in semileptonically tagged @xmath3-quarks .
this may significantly improve the precision which can be obtained .
detailed simulations will be necessary in order to determine how well the top quark charge can be measured using these methods .
we would like to thank c. ferreti , h. frisch , b. knuteson , t. lecompte , j. parsons , z. sullivan and j. womersley for stimulating discussions .
one of us ( u.b . ) is grateful to the fermilab theory group , where part of this work was carried out , for its generous hospitality .
this work has been supported in part by doe grant de - fg02 - 91er40685 , nsf grants phy-9600155 and phy-9970703 , and a nsf graduate student fellowship .
f. abe _ et al . _
( cdf collaboration ) , phys .
lett . * 74 * ,
2626 ( 1995 ) . s. abachi _ et al . _ ( d collaboration ) , phys .
lett . * 74 * , 2632 ( 1995 ) .
p. c. bhat ,
h. prosper and s. s. snyder , int .
j. mod .
* a13 * , 5113 ( 1998 ) .
d. chang , w.f .
chang , and e. ma , phys .
* d59 * , 091503 ( 1999 ) and phys .
* d61 * , 037301 ( 2000 ) .
b. grzadkowski and z. hioki , nucl .
b585 * , 3 ( 2000 ) and references therein .
g. velev , talk given at the top thinkshop@xmath199 , fermilab , november 2000 .
t. stelzer and w. f. long , comput .
commun . * 81 * , 357 ( 1994 ) .
h. murayama , i. watanabe and k. hagiwara , kek-91 - 11 ( 1991 ) .
a. denner and t. sack , z. phys .
* c46 * , 653 ( 1990 ) . t. affolder _
et al . _ , ( cdf collaboration ) , phys .
rev . * d63 * , 032003 ( 2001 ) .
b. abbott _ et al . _
( d collaboration ) , phys . rev . *
d58 * , 052001 ( 1998 ) .
d. abbaneo _ et al . _
( the lep electroweak working group ) , cern - ep-2001 - 021 ( february 2001 ) ; hep - ex/0103048 .
a. d. martin , r. g. roberts and w. j. stirling , phys .
* b387 * , 419 ( 1996 ) .
u. baur and e. l. berger , phys .
* d47 * , 4889 ( 1993 ) .
t. abe _ et al .
_ ( american linear collider working group collaboration ) , slac-570 _ resource book for snowmass 2001 , 30 jun - 21 jul 2001 , snowmass , colorado _ , hep - ex/0106055 , hep - ex/0106056 , hep - ex/0106057 and hep - ex/0106058 .
e. boos , hep - ph/0009100 , and references therein .
ccccc + & @xmath1 & @xmath109 & @xmath121 & + & contribution & contribution & contribution & total + tevatron , @xmath200 & 1.70 fb & 0.12 fb & 0.10 fb & 1.92 fb + tevatron , @xmath201 & 2.37 fb & 0.03 fb & 0.02 fb & 2.42 fb + lhc , @xmath200 & 33.3 fb & 5.78 fb & 4.86 fb & 44.0 fb + lhc , @xmath201 & 111.4 fb & 1.21 fb & 0.97 fb & 113.6 fb + + tevatron , @xmath200 & 0.36 fb & 0.26 fb & 0.13 fb & 0.75 fb + tevatron , @xmath201 & 0.52 fb & 0.07 fb & 0.03 fb & 0.62 fb + lhc , @xmath200 & 0.68 fb & 3.67 fb & 2.26 fb & 6.61 fb + lhc , @xmath201 & 2.25 fb & 0.87 fb & 0.59 fb & 3.71 fb + + tevatron , @xmath200 & 0.11 fb & 0.06 fb & 0.19 fb & 0.36 fb + tevatron , @xmath201 & 0.16 fb & 0.01 fb & 0.05 fb & 0.22 fb + lhc , @xmath200 & 0.15 fb & 1.07 fb & 3.63 fb & 4.85 fb + lhc , @xmath201 & 0.49 fb & 0.26 fb & 0.85 fb & 1.60 fb + .limits achievable at 95% c.l . for @xmath170 in @xmath188 at the tevatron and the lhc .
bounds are shown for the photon transverse momentum distribution in the phase space region defined by the @xmath1 selection cuts , and the @xmath122 and @xmath143 distributions of the combined radiative top decay regions .
the last row displays the combined limits from the three distributions .
the @xmath3-quark charge is given by @xmath162 .
the cuts imposed are described in sec .
ii . for the @xmath143 distribution , the @xmath122 cut of eq .
( [ eq : top4 ] ) has been replaced by @xmath151 gev . [ cols="^,^,^ " , ] | 15.pt we consider photon radiation in @xmath0 events at the upgraded fermilab tevatron and the cern large hadron collider ( lhc ) as a tool to measure the electric charge of the top quark .
we analyze the contributions of @xmath1 production and radiative top quark decays to @xmath2 , assuming that both @xmath3-quarks are tagged . with 20 fb@xmath4 at the tevatron , the possibility that the `` top quark '' discovered in run i is actually an exotic charge @xmath5 quark can be ruled out at the @xmath6 confidence level . at the lhc
, it will be possible to determine the charge of the top quark with an accuracy of about 10% . |
establishing an observational mapping between the properties of core - collapse supernovae ( ccsne ) explosions ( and related luminous outbursts ) and local populations of massive stars is key for constraining stellar evolution theory ( e.g. , langer 2012 ) .
arguably , the best link so far comes from the direct detections of red - supergiants with main - sequence masses @xmath6 m@xmath7 as progenitors of type iip supernovae ( e.g. , smartt 2009 , and references therein ) , the most common kind of ccsne .
however , direct detections of sn progenitors with higher masses ( @xmath8 m@xmath7 ) has been elusive ( e.g. , kochanek et al . 2008 ; smartt 2009 ) .
the detection of a very luminous star ( @xmath9 mag ) identified in pre - explosion images at the location of the type iin sn 2005gl provided the first direct evidence for a very massive h - rich star ( @xmath10 m@xmath7 ) that exploded as a luminous ccsn ( gal - yam et al .
2007 ; gal - yam & leonard 2009 ) .
these observations lend support for the connection made initially by smith et al .
( 2007 ) between one of the most luminous type iin sn 2006gy and very massive ( @xmath11 m@xmath7 ) luminous blue variable ( lbv ) stars like @xmath12 carinae . there is now mounting evidence of this connection ( e.g. , sn 2006jc , pastorello et al . 2007 ; sn 2010jl , smith et al .
2011a ; sn 1961v , kochanek et al .
2011 , smith et al .
2011b ) , which challenges massive stellar evolution theory in at least two important ways : 1 ) the most massive stars are not expected to be h - rich at the time when they explode as ccsne , at least with normal mass - loss prescriptions ; and 2 ) these stars have to loose up to a few solar masses of h - rich material in eruptions just before ( months to decades ) the core - collapse explosion in order to make them visible as type iin sne . on this last point of timing
strong eruptive mass - loss with the core - collapse explosion , there is some very interesting theoretical work , but its still very early days ( e.g. , woosley et al .
2007 ; arnett & meakin 2011 ; quataert & shiode 2012 ; chevalier 2012 ) .
luckily , in the last month nature has been kind and has provided the best case to date of a very massive star exploding as a bright core - collapse supernova : sn 2009ip .
this transient was discovered on 2009 aug .
29 by the chase survey in the outskirts of the galaxy ngc 7259 ( maza et al .
2009 ) and was initially given a sn name by the iau . however , detailed photometric and spectroscopic studies by smith et al .
( 2010 ) and foley et al .
( 2011 ) showed that it never quite reached supernova status .
it was actually a supernova impostor ( e.g. , humphreys & davidson 1994 , kochanek et al .
2012 ; with peak absolute mag .
@xmath13 , similar to @xmath12 carinae s great eruption , but significantly shorter timescale ) , in this case a massive lbv with estimated mass @xmath14 m@xmath7 ( constrained from pre - discovery imaging ) that had an eruptive mass - loss event . the catalina real - time transient survey ( crts ) discovered another outburst of comparable amplitude during 2010 ( drake et al .
2010 ) and a third outburst on 2012 july 24 ( drake et al .
2012 ) , the brightest so far .
the initial optical spectra obtained by foley et al .
( 2012 ) on aug .
26 showed narrow emission lines of h and he i , consistent with the spectra obtained in the 2009 outburst .
however , an unprecedented change of the spectrum was reported by smith & mauerhan ( 2012a ) .
15 and 16 they had detected very broad emission lines with p - cygni profiles , consistent with features observed in real type ii sne ; the massive lbv star had probably exploded as a supernova in real time ! another dramatic change in the spectral properties was reported by smith & mauerhan ( 2012b ) , by sep .
28 the broad p - cygni features had mostly disappeared , leaving behind narrow features characteristic of type iin sne .
all the details of their discovery and their follow - up observations can be found in mauerhan et al .
( 2012b ; hereafter m12b ) .
pastorello et al .
( 2012 ; hereafter p12 ) has also presented detailed follow - up observations of sn 2009ip and propose that the latest outburst might not be the final core - collapse explosion . in this paper
we present high - cadence optical observations of sn 2009ip that clearly resolve the brightening initially reported by brimacombe ( 2012 ) . in section
[ sec1 ] we discuss the observations , data reduction , and extraction of optical light curves using difference imaging photometry . in section [ sec2 ] we present an analysis of the different phases of the light curve . in section [ sec3 ] we present a discussion of the observed properties .
in section [ sec4 ] we present the conclusions of this work .
we adopt a distance of 20.4 mpc ( @xmath15 mag ) and galactic extinction of @xmath16 mag towards sn 2009ip ( m12b ) throughout the paper .
all the dates presented in this paper are ut .
imaging of the field of sn 2009ip was obtained by one of us ( j. b. ) between 2012 sep . 23.6 and oct . 9.6 from the coral towers observatory ( cairns , australia ) .
the data were collected from two different telescopes using different broad - band filters : 33-cm rcos ( with an @xmath17 filter ) and 41-cm rcos ( with an ir luminance filter , which has sharp blue cutoff at @xmath18 nm ) .
the ccd cameras used in both telescopes are identical 3k@xmath192k sbig stl6k , which give a total field of view of @xmath20 ( 1/pix platescale binned @xmath21 ) .
all the images were obtained using an exposure time of 900 sec .
a section of an ir image of sn 2009ip obtained on sep .
24 is shown in figure 1 .
we used the software maxim dl ( version 4.62 ) to complete the initial data reduction , which consists of bias subtraction , dark subtraction , and flat - fielding .
after visually inspecting all images and rejecting frames affected by bad tracking , we kept 85 ( @xmath17 ) and 118 ( ir ) frames for further photometric analysis .
we used the image subtraction package isis2 ( alard 2000 ) to extract the fluxes of sn 2009ip in individual frames following the steps described in hartman et al .
( 2004 ) . as reference images
, we used stacks of @xmath22 images with good seeing obtained on sep .
we carried out psf photometry with the daophot ii package ( stetson 1992 ) to subtract the sn flux and used the resulting sn - subtracted images as the final reference frames .
the sn fluxes obtained with isis2 in the @xmath17 and ir images were calibrated to standard @xmath17 and @xmath23 magnitudes , respectively , using data from the aavso photometric all - sky survey ( apass ) .
we obtained from apass calibrated sloan r and i magnitudes of 5 isolated stars in the field of sn 2009ip .
we converted these magnitudes in the sdss photometric system to standard @xmath24 ( johnson / kron - cousins ) photometry using the transformation equations obtained by r. lupton . in order to estimate the zeropoints
, we performed aperture photometry of the 5 local standard stars in the reference images using a 9radius aperture with the iraf task phot .
the resulting average zeropoints in the @xmath24 filters have a standard deviation of 0.02 mag .
the final photometry of sn 2009ip calibrated using these zeropoints is presented in table 1 and the light curves are shown in figure 1 .
the errors in each magnitude estimate include poisson errors from isis2 as well as the estimated standard error in the photometric zeropoints .
our photometry is in excellent agreement with the results presented in p12 , with mean differences during the period of observations of @xmath25 mag in @xmath17 and @xmath26 mag in @xmath23 .
the high cadence and number of observations allow us to clearly resolve the brightening of sn 2009ip , initially reported in brimacombe ( 2012 ) and prieto et al .
the optical light curves presented in figure 2 show well - defined phases : 1 ) approximately constant magnitude ( sep .
23 ) ; 2 ) very rapid brightening in a period of hours ( sep .
@xmath27 ) ; 3 ) turn - over to significantly slower brightening ( sep .
@xmath28 ) ; and 4 ) slow brightening , reaching peak magnitude ( sep . 30 @xmath29 oct . 9 ) .
we describe these phases in more detail below . between sep . 23.56 and 23.66
, the magnitude of sn 2009ip is consistent with being constant at @xmath30 mag ( @xmath2 ) .
however , between sep .
23.66 and 24.45 sn 2009ip brightened by @xmath31 mag and continued a very rapid brightening throughout the 6 hrs of continuous imaging obtained during the night of sep . 24 . in this period
, it brightened between @xmath32 ( sep .
24.45 ) and @xmath33 mag ( sep . 24.70 ) .
a linear fit to the @xmath23-band rise on sep .
24 gives a slope of @xmath34 mag day@xmath35 . between sep .
24.70 and 25.38 the sn brightened by @xmath36 mag , reaching @xmath37 mag ( @xmath38 ) .
the light curve turned over to a slower brightening on sep .
a linear fit to the data obtained during the night of sep .
25 gives a slope of @xmath39 mag day@xmath35 .
the turn - over can be seen clearly in both the @xmath17 and @xmath23-band photometric data obtained between sep .
25 and sep .
28 ( see figure 2 ) , reaching @xmath40 mag ( @xmath41 ) by sep . 28.47 . the latest part of the light curve between sep . 30 and oct .
7 shows a slow and steady brightening at a rate of @xmath42 mag day@xmath35 ( @xmath43 mag ) .
however , in our most recent images obtained on oct .
9.6 the sn faded by @xmath44 mag with respect to the images obtained on oct .
the magnitude of sn 2009ip on oct .
9.6 was @xmath45 mag ( @xmath4 ) .
the @xmath46 color is consistent with being constant ( @xmath47 ) during the period between sep . 25 and oct .
our photometric observations , obtained during a 16-day period between sep . 23.6 and oct . 9.6 show that the type iin sn 2009ip brightened by 3.7 mag in the optical to an absolute magnitude @xmath48 ( see also margutti et al .
2012b ; m12b ; p12 ) and it has reached peak magnitude around oct .
7 ( see figure 2 ) .
its peak luminosity is in the observed range of type ii sne ; in fact it is more luminous than most type iip sne ( e.g. , li et al .
the comparison to other type iin sne is less straightforward since the rates are lower and the samples are incomplete .
however , there are several well - studied type iin sne with peak absolute magnitudes around @xmath49 ( e.g. , kiewe et al .
2012 ; stritzinger et al . 2012 ; roming et al . 2012 ; mauerhan et al .
2012a ) . we can estimate the total radiated energy during the period of the observations presented here .
we need an estimate of the bolometric correction to convert the @xmath23-band luminosities to bolometric luminosities and we do this in two ways .
first , we use the observed constant @xmath46 color , corrected for galactic extinction , and fit a black - body to estimate the luminosity ratio @xmath50 and @xmath51 k. second , we use the longer wavelength coverage ( from near - uv to @xmath52-band ) of the swift observations presented in margutti et al .
( 2012b ) and fit a black - body to the extinction - corrected fluxes , obtaining @xmath53 and @xmath54 k. we integrate the @xmath23-band luminosities estimated from the extinction - corrected magnitudes .
assuming a constant bolometric correction during the 16-day observing period , we obtain a total ( optical ) radiated energy of @xmath55 erg and luminosities of @xmath56 erg s@xmath35 as of oct . 9 , depending on the correction used .
this is already comparable to the estimated radiated energy of the type iin sn 2011ht during its @xmath57 day plateau phase ( roming et al .
2012 ; mauerhan et al .
2012a ) .
the evolution of sn 2009ip since the discovery of its latest recorded eruption on 2012 jul .
24 ( drake et al . 2012 ) is quite remarkable ( figure 3 ; see also m12b and p12 ) , as it is unlike any other type iin sne studied to date .
the absolute magnitude was approximately constant at @xmath58 ( @xmath52 to @xmath23-bands ) for a period of @xmath59 days after jul .
24 , and even faded to @xmath60 ( at least between sep . 18 and sep . 22.5 , martin et al .
2012 ; margutti et al .
2012a ) before the rapid brightening that ( given our observations ) most likely started between sep . 23.7 and 24.5 .
if we assume the maximum bolometric correction estimated above as an upper limit and a plateau phase at @xmath61 for @xmath62 days , we constrain the total radiated energy during this period to be @xmath63 erg . however , we should note that the brightening could have started earlier than jul . 24 , in which case this could be a lower limit in the total radiated energy during this phase
. the spectroscopic evolution of sn 2009ip since its latest recorded eruption have also been unique .
foley et al .
( 2012 ) reported an optical spectrum obtained on aug .
26 , which showed a blue continuum with narrow ( @xmath64 km s@xmath35 ) emission lines of h and he i , similar to the spectra obtained during the 2009 eruption ( smith et al . 2009 ; foley et al . 2012 ) .
vinko et al .
( 2012 ) reported that an independent spectrum obtained on aug .
26 also showed broad p - cygni features with very fast velocities of @xmath65 km s@xmath35 , consistent with sn shock speeds .
these broad features seem to have persisted , and even strengthened , until ( at least ) sep .
23 ( m12b ) .
however , the high velocity features and p - cygni profiles had mostly gone away by sep .
26 - 27 , leaving behind strong , narrow ( @xmath66 km s@xmath35 ) emission features characteristic of many type iin sne spectra ( m12b ; burgasser et al . 2012 ; vinko et al . 2012 ; gall et al . 2012 ; p12 ) .
as shown in figure 3 , the transition in spectroscopic properties is coincident with the very rapid brightening observed in the optical light curves .
this , most likely , marks the start of a strong interaction between the fast sn shock and a dense csm environment likely formed from material ejected in the previous lbv outbursts , as proposed by m12b .
the timing of the rapid optical brightening is also coincident with the detection of x - ray emission from sn 2009ip ( campana & margutti 2012 ) , which is consistent with this picture . another possible signature of a strong ejecta - csm interaction is the decline of @xmath67 mag observed right before the spectral changes were seen and the light curve started to rise rapidly .
moriya & maeda ( 2012 ) propose that this is an unavoidable consequence of a fast sn shock breaking out of a dense csm .
figure 4 shows a zoom - in view of the evolution of the optical luminosity of sn 2009ip after sep .
23.5 , which is approximately when it started to transition to a type iin - dominated spectrum . for comparison
, we also show the optical luminosity evolution of three type iin sne from the literature that have relatively well - sampled light curves before and after peak brightness .
this sample includes the very luminous sn 2003ma ( rest et al .
2011 ) and sn 2006gy ( smith et al .
2007 ) , and the relatively low - luminosity sn 2011ht ( roming et al . 2012 ; mauerhan et al . 2012a
; but see humphreys et al .
2012 ) .
the very early rapid optical brightening of the type iin sn 2009ip is consistent with @xmath68 during a short 2-day phase , although this naturally depends on the @xmath69 used .
this is the evolution expected for an homologously expanding ( optically thick ) black - body photosphere at constant expansion velocity and effective temperature .
the luminous type iin sn 2003ma and sn 2006gy show relatively similar evolution in their optical luminosity at early times before peak .
in fact , smith & mccray ( 2007 ) modeled the early rise of sn 2006gy with c model .
sn 2011ht rises at a slower rate in the optical , although it was observed to rise much faster in the uv - range ( roming et al .
2012 ) . given the simplified assumptions that go into the @xmath70 scaling , we plan to explore more realistic models ( e.g. , moriya et al .
2012 ) to fit the early time light curve of sn 2009ip in a future publication . at later times , close to peak brightness ,
the optical luminosity and light curve shape of sn 2009ip becomes more similar to sn 2011ht than sn 2003ma or sn 2006gy .
mauerhan et al .
( 2012a ) proposed that sn 2011ht belongs to a growing sub - class of type iin sne ( type iin - p ; including also sn 2009kn , kankare et al .
2012 , and sn 1994w , sollerman et al . 1998 ) with long - lasting post - peak plateaus of @xmath57 days ( @xmath71 mag ) , and faint late - time decay slopes consistent with low @xmath72ni yield production ( @xmath73 m@xmath7 ) .
perhaps we are seeing something similar in the case of sn 2009ip ( but see p12 ) , although there are important differences in spectroscopic properties compared to the small group of type iin - p sne ( e.g. , sn 2009ip shows significantly hotter continuum and the narrow lines do not show p - cygni profiles ) .
an interesting related issue that comes out from the recent observations of sn 2009ip before the rapid brightening and fast transition to narrow - line dominated spectrum , is whether surveys might be missing a relatively low - luminosity phase ( at least compared to the main peak ) in which the broad sn shock velocities are revealed . given the typical depth of transient surveys , this is certainly possible ( e.g. , see fig .
17 in kiewe et al .
in fact , smith et al .
( 2011b ) proposed such an origin for the @xmath57 day plateau at @xmath74 mag observed before the main peak in the light curve of sn 1961v .
we have presented high - cadence optical photometric monitoring of the remarkable , young type iin supernova sn 2009ip using data obtained between 2012 sep . 23 and oct .
9 with @xmath75-@xmath76 meter aperture telescopes from the coral towers observatory ( cairns , australia ) .
using difference imaging analysis , we obtain precise @xmath17 and @xmath23-band light curves and are able to resolve well - defined brightening phases .
in particular , we see a very rapid brightening early on ( 0.5 mag in 6 hr ) that quickly turns over after a couple of days .
the sn brightened between absolute magnitude @xmath77 and @xmath78 in 16 days .
the changes that we observe in the light curves are correlated with the reported spectroscopic changes and the x - ray detection , and , most likely , mark the start of a strong interaction between the fast sn ejecta and the dense csm formed from material that was ejected in the pre - supernova lbv eruptions observed since 2009 .
our study highlights the importance of organized , high - cadence observations by amateur astronomers with small - aperture telescopes and the impact they can have in transients and sne research , in this case with direct important connections to massive stellar evolution .
we thank s. de mink , o. graur , j. mauerhan , r. quimby , and a. rest for stimulating discussions about sn 2009ip , and l. watson for detailed comments on an earlier version of this manuscript . j. l. p. acknowledges support from a carnegie - princeton fellowship .
lcccc sep 24.536400 & 195.040909 & 16.549 & 0.047 & @xmath17 + sep 24.552130 & 195.056641 & 16.494 & 0.045 & @xmath17 + sep 24.567801 & 195.072327 & 16.382 & 0.040 & @xmath17 + sep 24.583472 & 195.087997 & 16.403 & 0.043 & @xmath17 + sep 24.599144 & 195.103668 & 16.356 & 0.044 & @xmath17 + sep 24.614826 & 195.119370 & 16.256 & 0.039 & @xmath17
+ sep 24.633634 & 195.138153 & 16.211 & 0.040 & @xmath17 + sep 24.644965 & 195.149475 & 16.259 & 0.043 & @xmath17 + sep 24.656285 & 195.160797 & 16.256 & 0.040 & @xmath17 + sep 24.667593 & 195.172134 & 16.250 & 0.038 & @xmath17 + & & & & @xmath17 + | recent observations by mauerhan et al .
have shown the unprecedented transition of the previously identified luminous blue variable ( lbv ) and supernova impostor sn 2009ip to a real type iin supernova ( sn ) explosion .
we present high - cadence optical imaging of sn 2009ip obtained between 2012 ut sep . 23.6 and oct . 9.6 , using @xmath0 meter aperture telescopes from the coral towers observatory in cairns , australia .
the light curves show well - defined phases , including very rapid brightening early on ( 0.5 mag in 6 hr observed during the night of sep .
24 ) , a transition to a much slower rise between sep . 25 and sep . 28 , and a plateau / peak around oct . 7 .
these changes are coincident with the reported spectroscopic changes that , most likely , mark the start of a strong interaction between the fast sn ejecta and a dense circumstellar medium formed during the lbv eruptions observed in recent years . in the 16-day observing period sn 2009ip brightened by 3.7 mag from @xmath1 mag on sep .
23.6 ( @xmath2 ) to @xmath3 mag ( @xmath4 ) on oct . 9.6 , radiating @xmath5 erg in the optical wavelength range .
currently , sn 2009ip is more luminous than most type iip sne and comparable to other type iin sne . |
following einstein s insight that gravity is encoded into the geometry of spacetime , quantum gravity aims at providing realizations of quantum geometry . in this task ,
one key technical and conceptual challenge is to reconcile the regularizations required in quantum field theory with the diffeomorphism symmetry which underlies general relativity . indeed ,
a number of approaches employ discretizations as regulators , which is the case for instance of regge calculus @xcite , and in this class of theories , where one attempts to represent geometrical data on a triangulation , diffeomorphism symmetry is generically broken @xcite . in light of this ,
a very important result is therefore the construction of a continuum notion of quantum geometry , which was achieved in the context of loop quantum gravity @xcite ( lqg hereafter ) by ashtekar , isham , and lewandowski @xcite .
the so - called ashtekar lewandowski ( al ) representation provides a hilbert space representation of the kinematical algebra of observable of the full _ continuum _ theory .
this kinematical algebra of observable encodes the intrinsic and extrinsic geometry of a spatial hypersurface into holonomies measuring curvature ( of the ashtekar barbero connection @xcite ) and fluxes measuring the spatial geometry .
most importantly , this kinematical setup allows to deal successfully with spatial diffeomorphisms .
indeed , there exists a fully kinematical hilbert space describing geometry in a coordinate - dependent way , and on which the diffeomorphisms act unitarily .
the fact that the diffeomorphisms act unitarily also allows to define a hilbert space of spatially diffeomorphism - invariant states , and , quite noticeably , this task has so far only been achieved in the ( ashtekar ) connection formulation of general relativity @xcite .
the kinematical hilbert space allows for a rigorous definition of the quantum dynamics in the form of hamiltonian constraints @xcite , and one can hope to construct the physical hilbert space , which would incorporate spacetime diffeomorphism - invariant states , from the solutions to these constraints . in the present article ,
building upon the earlier work @xcite , we construct an alternative realization of a continuum quantum geometry , which is unitarily inequivalent to the ashtekar lewandowski representation .
we hope that this new framework will make the description of states describing configurations with macroscopic geometry much easier .
the reason for this expectation is that the new representation which we are constructing here supports states which are peaked on an almost everywhere flat connection ( we therefore call this representation the bf representation , since the bf vacuum is peaked on a flat connection ) .
curvature has only distributional support on defects , which brings us much nearer to regge s proposal @xcite of approximating general relativity by a very dense gas of defects in an otherwise flat geometry .
as explained in more detail in the overview given in section [ overview ] , our construction leads to a continuum hilbert space which supports arbitrarily many excitations in the form of defects .
in fact , both the al representation and the bf representation can be interpreted in this way . for the al representation , it is the defects themselves which generate ( non - degenerate ) geometry , and therefore a macroscopic geometry corresponds to a highly excited state . for the bf representation however ,
the states have ( almost everywhere ) maximal uncertainty in spatial geometry since they are peaked on the conjugated variable , namely on flat connections .
there are numerous proposals to approximate the dynamics of general relativity by mostly flat geometries with defects , either in the classical theory @xcite , or in the context of lqg @xcite .
another related class of approaches is the combinatorial quantization of flat connections in three spacetime dimensions @xcite . in this work ,
we construct for the first time a hilbert space carrying a representation of a continuum observable algebra and supporting states which are peaked on almost everywhere flat connections .
this continuum construction has important consequences .
in particular , it requires the compactification ( or the exponentiation ) of the fluxes , and with it the introduction of a discrete topology for the holonomy group parameters .
this in turn changes the properties of , e.g. , the spectra of observables like the area operator .
our construction of the continuum hilbert space is done via a so - called inductive limit . as is explained in more detail in section [ overview ] ,
such a constructions allows one to keep the cake and to eat it too . for most purposes ,
it is sufficient to deal with a discretization .
however , all the hilbert spaces describing states restricted to such discretizations are embedded into a continuum hilbert space .
the properties of observables are therefore changed further due to redefining ( as compared to the al representation ) the way in which the observables on hilbert spaces associated to discretizations are related to the observables on the continuum hilbert space . for the bf representation ,
this relation is based on a geometric coarse - graining of the fluxes , which allows to address for example the staircase problem appearing for geometric operators in the al representation .
this work is divided in two main parts .
we start in section [ sec : configuration space ] by providing a characterization of the ( cotangent ) space of flat connections on manifolds with a fixed number of defects , where the positioning of the defects is prescribed by a choice of triangulation .
we then define in section [ sec : inner product ] an inner product ( compatible with the inductive limit construction which comes afterwards ) which leads to a discrete topology on the group .
this discrete topology on the group implies that dealing with gauge invariance is far less trivial than in the al case .
we bypass this difficulty by working with the space of almost gauge - invariant wave functions , and employ a group averaging for the remaining global adjoint action . in section [ sec : quantization algebra ] , we construct a representation of holonomies and ( exponentiated ) fluxes which are supported by the triangulation .
this already allows us to discuss the spectra of these operators , and also that of the area operator , which we do in section [ sec : area ] . in the second main part of this work
, we construct the continuum hilbert space via an inductive limit .
this first requires the definition of embeddings that map hilbert spaces associated to coarser triangulations into hilbert spaces associated to finer triangulations .
we therefore develop a number of technical results in section [ sec : refinement ] , which allow in particular to obtain a convenient splitting of the observables into coarser observables and finer observables . in section [ sec : inductive limit ] , we describe the construction of the inductive limit hilbert space .
we then discuss how to relate the operators on the various hilbert spaces in order to define a notion of continuum operators in sections [ sec : extop ] and [ contop ] . finally , we close with a discussion of our results in section [ sec : discussion ] .
the appendices include a more detailed elaboration on the configuration space of flat connections , a number of proofs of theorems needed in the main text , as well as the construction of the ( intertwiner- ) spin representation for the wave functions , which replaces the usual spin network states of the al representation .
in order to construct the _ continuum _ hilbert space , we proceed in the same way as for the al representation , namely by considering an inductive limit of hilbert spaces .
let us explain why this construction is very natural in a background - independent context .
this will also allow us to explain how the new realization of quantum geometry presented here can be seen as one member of a potentially much bigger class of quantum geometries @xcite .
we usually organize hilbert space representations into a vacuum and excitations , where the excitations are generated by applying the observable algebra to the vacuum state .
the vacuum states we are dealing with for both the al and the new representation are states which are sharply peaked on a certain configuration , as opposed to the gaussian states of free field theory for which one needs to specify a background metric . for the al representation ,
one has a vacuum which is sharply peaked on a totally degenerate spatial geometry , i.e. in which all flux observables annihilate the vacuum . for the representation constructed in this work ,
the vacuum state is very different : it is sharply peaked on a ( globally and locally ) flat connection .
in fact , such states which are sharply peaked on a flat connection can be described by ( first class ) constraints arising from a topological field theory ( these constraints play also an important role in the classical theory @xcite ) . for the new representation presented here ,
this topological field theory is given by bf theory @xcite , which is also the theory underlying the construction of spin foam models @xcite .
we will therefore refer to the new representation as the bf representation .
the topological field theory for the al representation is a trivial one , and coincides with the strong coupling limit of yang
mills theory , whereas bf theory arises from a zero coupling limit @xcite .
the topological field theory in itself does however not provide room for local excitations , and therefore does not give a representation of the observable algebra which is supposed to be generating such excitations . in order to add excitations to the theory
, one needs to allow for the presence of defects , e.g. in the form of a violation of the flatness constraints in the bf case , or of a violation of the flux constraints in the al case .
therefore , the notion of what constitutes a `` defect '' strongly depends on the choice of vacuum . in the case of the al representation ,
a `` defect '' actually means an excitation which allows to have at least some of the spatial geometrical operators with non - vanishing values .
this would rather be an `` anti - defect '' of geometry if one understands a degenerate geometry as being highly defected .
defects play also an important role in condensed matter applications @xcite . in this case , one wishes to describe localized excitations which are separated by very large distances so that they do not interact with each other .
therefore , the precise values of the distances , i.e. the metric geometry , does not actually matter .
this justifies a posteriori the use of the vacuum ( or the physical state ) of a topological field theory : it ensures that only relations of topological nature between the defects actually matter ( for instance how two line defects wind around each other ) .
furthermore , it explains why frameworks for describing defects are applicable to quantum gravity , since the metric properties are then not encoded in a background , but in the defect configurations themselves .
a second property which one usually assigns to defects is a discreteness for their `` charge '' labels . in condensed matter , this arises since one needs finite energy gaps between the vacuum and the excitations described by the defects , so that there is a notion of adiabatically moving the defects around .
interestingly , there is a similar discreteness of the charges describing the defects in the framework of quantum geometry .
this arises however due to a different reason , namely in the way in which the continuum limit is constructed . to construct the continuum limit
, one starts with a family of hilbert spaces where each hilbert space describes a fixed number of defects . for a diffeomorphism - covariant description
, one does not only fix the number of defects , but also their positions .
this is achieved by embedding a discrete structure , e.g. a graph or a triangulation , into the spatial manifold .
defects are then confined to this discrete structure , for instance line defects to the links of the graph or to the edges of the triangulation .
the degrees of freedom are given by the values of the charges for the defects .
the values can also describe a vanishing charge , which is then equivalent to having no defect .
this is an important point since it allows to embed states realized in a hilbert space based on e.g. a coarse triangulation into a hilbert space based on a refinement of this triangulation . for the continuum theory ,
one however wishes to allow for an arbitrary number of defects ( at arbitrary positions ) .
one can construct such a continuum hilbert space as an inductive limit of hilbert spaces . a key point in this construction
is the notion of refinement for the discrete structures carrying the defects on the one hand , and for the states on the other hand .
let us consider the example of line defects confined to a graph embedded into the manifold .
one possible refinement of a graph is to add links to it .
the corresponding state for the refined graph is then defined by transposing over all the charge labels from the coarse state , and by assigning the charge labels describing a vanishing defect for the new links . in other words ,
one puts the new additional degrees of freedom into a vacuum state @xcite . in order to define the inductive hilbert space structure , the refinement operator acting on the states needs to be isometric ( with respect to the inner product defined on the hilbert spaces with a fixed discrete structure ) .
in particular , the vacuum state where all the charges describe vanishing defects should have a finite norm .
moreover , what happens ( in both the bf and the al representation ) is that the charges label an orthonormal basis for the hilbert spaces associated to a fixed discrete structure .
therefore , even if the charge labels are continuous ( as it is the case with the bf representation , where the labels are elements of the ( lie ) group ) , we will have to equip the space of charge values with a discrete topology . if we see the charge labels as momentum space , then a discrete topology on momentum space corresponds to a compactification of the configuration space .
this compactification can be understood in the following way for the two representations : * in the al representation , the configuration space associated to a fixed discrete structure is given by group holonomies along the links of embedded graphs , and therefore corresponds to @xmath0 , where @xmath1 is a compact lie group ( usually @xmath2 ) , and @xmath3 denotes the number of allowed defects , which in this case coincides with the number of links of the graph .
the inductive limit of hilbert spaces amounts to a projective limit for the configuration space , in which compactness is preserved ( see for instance @xcite ) .
the space of charge labels is constructed through the group fourier transform and given by representation labels ( and magnetic indices ) .
this space does indeed naturally carry a discrete topology ( for a compact group @xmath1 ) .
* in the bf representation , the configuration variables are actually the fluxes and [ intertwinerspin ] . ] , which correspond to the dual of the lie group .
we therefore need to consider a compactification of this space .
in fact , for the group @xmath4 the ( pontryagin ) dual is given by the abelian group @xmath5 , which can be bohr - compactified .
this in turn leads to a discrete topology on @xmath4 . for @xmath6 , it is however easier to start with a discrete topology , and to then derive the compactification of the dual space via a ( generalized ) group fourier transform .
this is described in appendix [ spinrepsu2 ] .
therefore , we see that the inductive limit construction leads to a compactification of the underlying configuration spaces .
this also explains the difficulty to define a projective limit hilbert space ( in the al representation ) for a non - compact group like @xmath7 , since this latter can not easily be compactified , one can use the bohr compactification @xcite .
it might actually be easier to compactify the dual of @xmath7 , which would then allow for a bf representation . ] .
the compactification of the fluxes has important repercussions for the properties , and in particular the spectrum , of the observables . due to the discrete topology on the ( lie ) group
, the fluxes can not act anymore as ( lie ) derivatives like in the al representation .
instead , one has to replace the fluxes with translation operators labelled by a translation group element @xmath8 . due to the compactification
, the spectrum of these translation operators is bounded . however , depending on the choice of @xmath8 , the spectrum can be either discrete or continuous .
the continuous spectrum arises if the action of the translation operator on the group is ergodic .
we will provide a more detailed discussion of these features in section [ sec : area ] .
a ( bohr ) compactification is also employed in another context in loop quantum cosmology ( lqc ) @xcite . in this case ,
one compactifies @xmath9 , which describes the momentum conjugated to the scale factor ( this momentum @xmath10 is proportional to the connection degree of freedom ) .
this then leads to a discrete topology for the scale factor .
note that in lqc ( point ) holonomies are compact while the conjugated momentum discrete , which is the reversed situation as compared to the bf representation .
it is also well - known that compactifications lead to non - separable hilbert spaces .
let us comment more on this issue , and in particular on the fact that there are now two different sources of non - separability .
* the first source of non - separability arises from the inductive limit hilbert space construction based on _ embedded _ discrete structures .
this can be interpreted as introducing a discrete topology for the underlying spatial manifold .
this is also mirrored in the interpretation of defects in condensed matter as charges which lie at large distance from one another .
here , we declare for instance any point ( in the case in which the defects are point - like ) of the underlying manifold to be separated from any other point ( mathematically , these points define open sets ) , which gives rise to the discrete topology .
+ one can however expect that this kind of non - separability will be lifted ( to a large extent ) if one constructs the spatial diffeomorphism - invariant hilbert space
. indeed , diffeomorphisms act by changing the embeddings of the discrete structures , and therefore the positioning of the defects .
for a diffeomorphism - invariant description only the relative ( topological ) positions of the defects matter . * in the bf representation , we encounter a second kind of non - separability , which results from the discrete topology on the space of charge labels .
this non - separability will not be lifted by implementing spatial diffeomorphism symmetry .
there is however the possibility that it can be lifted for the physical hilbert space if one includes a positive cosmological constant ( at least in the euclidean theory ) . in this case
, the large spin @xmath11 labels in the spin representation might become irrelevant : for instance , the area eigenvalues first grow with growing spin , and then start ( depending on the value of @xmath8 ) to oscillate .
therefore , very large spins are in a sense not needed since they lead to similar area eigenvalues as the smaller spins .
+ one can also introduce a quantum deformation of the group at root of unity , i.e. work with @xmath12 .
in this case , the cutoff on the spin is implemented explicitly .
the charge labels are then described by the drinfeld center of the @xmath12 module category , which has only finitely many objects @xcite . in three spacetime dimensions ,
the corresponding vacuum is given by the turaev
viro state sum model @xcite and describes ( euclidean ) gravity with a positive cosmological constant .
future work @xcite will be concerned with constructing a continuum hilbert space representation based on quantum groups .
in this section we briefly summarize the setup of our construction , and present the notations and notions which are used throughout the rest of the article .
we refer the reader to the companion paper @xcite for more details .
let @xmath13 be a @xmath14-dimensional closed ( i.e. compact and without boundary ) manifold , which we choose to be real , analytic , orientable , and path - connected .
we will be interested in particular in the cases @xmath15 and @xmath16 , which we treat simultaneously .
the manifold @xmath13 can be thought of as a spatial boundary , and our aim is to construct states corresponding to @xmath14-dimensional boundary quantum geometries on it .
for this , we consider a principal @xmath1-bundle @xmath17 , where the gauge group @xmath1 is an arbitrary compact lie group whose lie algebra will be denoted by @xmath18 . if @xmath1 is simply - connected ( which is the case for @xmath2 , the gauge group of interest in this work ) , since we consider the case @xmath19 it is always possible to pick a global trivialization of @xmath20 , which we do for simplicity .
the space @xmath21 of smooth @xmath1 connections is then affine over @xmath22 , the space of @xmath18-valued one - form fields over @xmath13 , and the group of local gauge transformations is given by @xmath23 .
on the phase space of lqg , a connection @xmath24 is canonically conjugated to a @xmath25-valued vector density @xmath26 ( the electric field ) of weight + 1 .
if we write @xmath27 and @xmath28 , where @xmath29 are coordinates on @xmath13 and @xmath30 forms a basis of @xmath18 , the non - vanishing poisson brackets are given by [ firstequ ] a^i_a(x),e^b_j(y)=^b_a^i_j^d(x , y ) .
following the logic of lqg , we are going to consider holonomies of the connection and fluxes of the electric field as our elementary variables .
however , instead of assigning this data to arbitrary oriented embedded ( dual ) graphs and @xmath31-dimensional submanifolds , as is done in the usual construction of the holonomy - flux @xmath32-algebra , here we adopt a simplicial point of view and focus rather on the triangulation itself .
we now explain the type of discrete data which is relevant for our construction .
we consider a fiducial riemannian metric on the manifold @xmath13 , and the infinite set of embedded ( or geometric ) triangulations @xmath33 defined by this metric . in both @xmath15 and @xmath16 dimensions ,
these embeddings refer to the vertices of the triangulations .
the edges and ( in @xmath16 ) the triangles of a given triangulation are defined respectively as geodesics and minimal surfaces with respect to the auxiliary metric .
we furthermore assume that the triangulations are fine enough so that these geodesics and minimal surfaces are unique . to each triangulation
@xmath33 we assign a dual graph @xmath34 , which is the one - skeleton of the cellular complex dual to @xmath33 .
this dual graph is such that its nodes @xmath35 are dual to the @xmath14-dimensional simplices @xmath36 of @xmath33 , while its links @xmath37 are dual to the @xmath31-dimensional simplices @xmath38 of @xmath33 . for each link
we choose a fiducial orientation , and denote by @xmath39 and @xmath40 the source and target nodes . to each pair given by a triangulation and its dual graph
we then assign a root , which is a choice of a preferred @xmath14-dimensional root simplex @xmath41 , or equivalently ( by duality ) a choice of a preferred root node @xmath42 .
this in turn defines a rooted geometric triangulation . in order to describe the behavior of this root under the refining operations , we further equip the root simplex with a flag structure , which consists simply in a choice of nondisjoint lower - dimensional simplices @xmath43 . with these structures
, we can finally consider the infinite set of flagged rooted geometric triangulations of a given manifold @xmath13 . on this set
, we now choose the refinement operations to be the alexander moves .
these alexander moves refine a given triangulation by subdividing a @xmath44-dimensional simplex , where @xmath45 , with the addition of a new ( embedded ) vertex . as explained in @xcite ,
if the root of a triangulation is subdivided by a move , the flag structure defines in a canonical manner a new root for the refined triangulation . a geometric triangulation @xmath46 is said to be finer than a geometric triangulation @xmath33 if it can be obtained form @xmath33 by a ( finite ) sequence of refining alexander moves , and if its vertices after the moves have the same coordinates as those of @xmath33 .
we will denote this relation by @xmath47 .
in oder to unambiguously specify paths @xmath48 in a graph @xmath34 and to obtain a characterization of the fundamental cycles of this graph ( and to perform a gauge - fixing ) , it is convenient to introduce ( maximal spanning ) trees . a maximal or spanning tree @xmath49 in a graph @xmath34 is a connected subgraph of @xmath34 which does not contain any closed cycles and includes all the nodes of @xmath34 . in the rest of the paper we use
` tree ' as short form for maximal connected tree .
the links composing a tree are called branches and denoted by @xmath50 , while the links of the graph that are not in the tree are called leaves and denoted by @xmath51 .
although the leaves correspond to the links in @xmath52 , by abuse of language we will refer to the leaves of the tree .
the set of leaves in a tree will be denoted by @xmath53 .
a rooted tree with root @xmath42 is a tree where a preferred node is identified and called the root .
there is a one - to - one correspondence between the leaves of a tree and the fundamental cycles of the underlying graph . for a rooted tree , one way to characterize these cycles
is to associate them with paths @xmath54 labeled by the leaves in @xmath53 .
a path @xmath54 starts at the root , goes along the branches of the tree until the source node of the leaf @xmath51 , then goes along @xmath51 , and back to the root along branches .
the cardinality of the set of fundamental cycles is independent of the choice of tree .
explicitly , if we denote by @xmath55 the number of nodes in the graph @xmath34 and by @xmath56 the number of links , then any given tree has @xmath57 branches and @xmath58 leaves ( and fundamental cycles ) .
as explained in the introduction , to construct the full continuum quantum theory we will proceed in two steps . in this first part
, we are going to define the quantum theory on a fixed triangulation .
for this , we will first focus on the classical connection configuration space , then on the inner product and hilbert space structure built upon it , and finally on the quantization of the algebra of holonomies and fluxes .
this will enable us in particular to discuss the properties of the area operator .
in the second part we will then show the existence of a continuum ( inductive limit ) hilbert space .
in this section , we focus on the connection degrees of freedom and describe the configuration space of smooth @xmath1 connections on the spatial manifold @xmath13 . in the al framework , the quantum theory is built over a suitable generalization of @xmath21 which is the space @xmath59 of generalized connections . in the present context however , as we shall see , by configuration space we mean the space of classical states which reflect the geometrical properties of the states of our ( yet - to - be built ) quantum theory , which turns out to be the moduli space of flat connections on the spatial manifold @xmath13 with certain defects . ]
over which we are going to build the quantum theory .
the bf vacuum is a state locally and globally peaked on flat connections .
this property can be encoded in the triviality of the holonomies of the connection around the faces of the dual graph @xmath34 ( more precisely , along the fundamental cycles of the graph ) .
excitations over this vacuum locally shift the holonomies away from the identity , thereby creating curvature defects which can be thought of as being carried by the @xmath60-dimensional simplices ( the hinges ) of the triangulation .
for this reason , we are interested in connections which are flat appart from possible conical singularities along the hinges .
if we denote by @xmath61 the @xmath60-skeleton of the triangulation @xmath33 , we are therefore led to considering the space of flat connections on @xmath62 . as is well - known @xcite , the moduli space of flat connections modulo gauge transformations on a compact and connected manifold @xmath13 is a finite - dimensional orbifold which admits several different ( although equivalent ) mathematical descriptions . in particular
, it is common to describe the quotient _ 0 a|f(a)=0/ in terms of holonomies ( i.e. group homomorphisms ) as [ without defects ] _ 0=(_1(),g)/g , where @xmath1 acts on the space @xmath63 of homomorphisms by diagonal adjoint action .
this defines the space of locally flat connections for a compact and connected manifold @xmath13 without defects .
details about the construction of the first fundamental group @xmath64 can be found in appendix [ appendix : pi1 ] . in the present work ,
since we are interested in flat connections on the defected manifold @xmath62 , we first have to worry about the fact that this latter is not compact . in order to avoid the inconvenience of working with infinite triangulations
, one can resort to the following construction : first , choose a flat metric on the simplices of @xmath33 ( which can be the auxiliary metric on @xmath13 ) and let , for some small @xmath65 , @xmath66 be the open ball of radius @xmath67 .
then , the set _ ( d-2)^_x_(d-2)b_(x ) , which represents a `` blown - up '' version of @xmath61 , is open in @xmath13 , and if @xmath67 is small enough one has that @xmath62 is homotopy - equivalent to the new manifold @xmath68 .
in fact , one has that @xmath69 is a deformation - retract of @xmath62 .
as such , these two spaces have isomorphic fundamental groups , and we can define our configuration space to be _ ( _ 1(),g)/g .
the question is now how to characterize @xmath70 . to this end
we note that the graph @xmath34 , which is the one - skeleton of the simplicial complex dual to the triangulation @xmath33 of @xmath13 , is a deformation - retract of @xmath71 . as a consequence ,
their first fundamental groups are isomorphic , i.e. @xmath72 .
we are therefore led to considering the first fundamental group @xmath73 of the graph @xmath34 dual to the triangulation @xmath33 .
the first fundamental group @xmath73 of a graph @xmath34 is a free group , and is generated by a set of fundamental cycles of the graph @xcite ( we refer the reader to appendix [ appendix : pi1defects ] for more details ) .
as noted in section [ sec : setup ] , a set of fundamental cycles can be obtained by choosing a maximal spanning tree @xmath49 ( which here in addition will be rooted ) in @xmath34 .
the links of @xmath34 which are not part of the tree are called leaves and denoted by @xmath51 . the set @xmath53 of leaves with respect to a given choice of tree @xmath49 defines a set of fundamental cycles @xmath54 .
such a fundamental cycle @xmath54 starts at the root , goes along the branches of the tree until the source node of the leaf @xmath51 , then goes along @xmath51 , and back to the root along branches . by definition , a set of fundamental cycles obtained by a choice of tree
does freely generate the first fundamental group @xmath73 .
in other words , if we denote by @xmath54 the cycle associated to the leaf @xmath51 and choose a numbering @xmath74 of the leaves , we have that _ 1 ( ) = _ 1, ,_|| .
therefore , the space @xmath75 is parametrized by the set of group - valued maps @xmath76 , and by associating to each cycle @xmath54 a group element @xmath77 we can write that ( _ 1(),g ) = \{g_1, ,g_|||g_ig}. from this space , we then obtain the space of locally flat connection by dividing out by the diagonal adjoint action of the group on @xmath78 , i.e. _ \{g_1, ,g_|||g__ig}/g .
as we will now explain , this space in turn agrees with the space of gauge - invariant ( or gauge - fixed ) holonomies on the graph @xmath34 .
since the @xmath79-tuple @xmath80 determines the holonomies along the fundamental cycles of the graph , it can be used to reconstruct the holonomies along all the cycles ( starting and ending at the root ) in @xmath34 .
a ( suitable ) space of functions of these cycle holonomies therefore gives the space of functions of the graph connection which are invariant under gauge transformations at all the nodes except the root .
let us explain how this comes about .
a graph connection is given by an assignment of group elements @xmath81 to the links @xmath82 of the graph , and the action of a gauge transformation with parameter @xmath83 on @xmath81 factorizes at the nodes and gives u_nh_l = u^-1_l(1)h_lu_l(0 ) .
given a closed cycle @xmath84 with @xmath85 starting at the root and @xmath86 ending at the root , the cycle holonomy @xmath87 is therefore invariant under gauge transformations at all the nodes except at the root . instead , under gauge transformations at the root , it transforms with the adjoint action [ rootgaugetrafo1 ] g^|n|\{u_n}g_=u^-1_r g_u_r . by choosing a tree ,
the space of holonomies which are invariant under non - root gauge transformations can again be characterized by the association of group elements to the leaves of the graph . as explained above , for ever leaf
@xmath51 we can consider paths @xmath54 describing the fundamental cycles by starting at the root , going along the tree until the source of @xmath51 , then going along @xmath51 , and back to the root along the tree . to every leaf we can therefore associate a group element [ hcycle ] g_h_e^1 = g_r(1)^-1h_g_r(0 ) .
any other arbitrary cycle holonomy can be obtained by composing the holonomies @xmath77 of the fundamental cycles .
the fundamental cycle holonomies therefore parametrize the space of graph connections which are invariant under non - root gauge transformations .
furthermore we can use the tree to perform a ( partial ) gauge - fixing . given any graph connection @xmath88 there exist a unique gauge transformation @xmath89 , acting on all nodes except the root , such that @xmath90 for all the links @xmath50 which are branches of the tree . here
we use @xmath91 to denote an equality which holds with a particular gauge - fixing . in this gauge
, the holonomy along a fundamental cycle reduces to the holonomy associated to the leaf @xmath51 , i.e. @xmath92 . thus @xmath93 parametrizes the space of almost gauge - invariant graph connections , that is connections invariant under all gauge transformations except the ones at the root .
we can therefore conclude that @xmath78 describes the space of almost gauge - invariant connections on @xmath34 , as well as the space of almost gauge - invariant locally flat connections on the manifolds with defects @xmath71 , which we will denote by @xmath94 .
in this section we introduce the inner product and hilbert space structure on a fixed triangulation . for this
, we first focus on states which are gauge - invariant everywhere appart from the root , and then use the technique of group averaging in order to account for these residual gauge transformations . our goal is to construct a hilbert space @xmath95 for which the space of almost gauge - invariant graph connections @xmath94 serves as the configuration space , as it would be the case for example in ( at least one approach to ) the quantization of chern simons theory . ] .
the reason for working with @xmath94 instead of the full gauge - invariant space @xmath96 is that the configuration space @xmath78 is much simpler to handle than the orbifold @xmath97 .
we will therefore first consider @xmath94 , and then impose invariance under the residual gauge transformations at the root via a group averaging procedure .
as discussed in section [ sec : configuration space ] , the configuration space @xmath94 is parametrized by a set of group elements in @xmath78 , and the identification of the leaves @xmath51 implicitly means that one has made a choice of tree @xmath49 .
we therefore consider some space @xmath98 of functions as a candidate for our hilbert space @xmath95 , and consider states of the form |\{g _ } ( g_1, ,g_|| ) . here
we can interpret the group elements @xmath77 as giving the holonomies associated to the fundamental cycles @xmath54 of the graph @xmath34 . alternatively ,
if we use the tree @xmath49 to gauge - fix the branch holonomies to the identity , since in this gauge we have @xmath92 , we can interpret the group elements @xmath77 as the ( edge ) holonomies associated to the leaves .
let us now propose an inner product on the space @xmath98 .
our choice for this inner product is motivated by the inductive limit hilbert space construction which will be carried out in the second main part of this work .
as we will see , this construction requires to isometrically embed hilbert spaces based on coarser triangulations @xmath33 into hilbert spaces based on finer triangulations @xmath46 .
recall that for a finer triangulation @xmath46 with @xmath99 , the associated manifold @xmath100 has more defects than the manifold @xmath62 , and the dual graph @xmath101 has more independent cycles than the dual graph @xmath34 .
this means that the space @xmath102 of almost gauge - invariant holonomies is isomorphic to @xmath103 with @xmath104 .
now , the coarser configuration space @xmath94 can be regained from the finer configuration space @xmath102 by imposing constraints @xmath105 ( where the index @xmath106 has range @xmath107 ) .
these constraints impose that the `` finer '' holonomies , i.e. holonomies of cycles around the additional defects in @xmath100 , are trivial ( this will be explained in much more detail in section [ sec : refinement ] ) .
the presence of these constraints means that states in the finer hilbert space @xmath108 which arise as the embedding of states in the coarser hilbert space @xmath109 are sharply peaked on vanishing `` finer '' holonomies .
these states need to have a finite norm as otherwise the embedding would not be isometric ( and the inner product would not be cylindrically consistent , which would prevent the inductive limit construction of the continuum hilbert space from existing ) .
furthermore , the motivation for this work is to construct a representation based on a vacuum state which is peaked on locally and globally flat connections , and this vacuum state should of course also have a finite norm . if the gauge group @xmath1 is a finite group , the inner product provided by the discrete measure ( which agrees in this case with the haar measure ) leads to the required finiteness .
however , if @xmath1 is a lie group ( even a compact one ) , the states of interest will have infinite norm .
we therefore need to consider an appropriate regularization procedure .
our regularization procedure goes as follows .
first , we start with an auxiliary inner product @xmath110 , together with a family @xmath111 of regulator states having a finite norm with respect to this auxiliary inner product for finite @xmath112 .
second , we need a family @xmath113 of regulator states approaching the ( locally and globally ) flat vacuum state @xmath114 when @xmath115 . with this , we can then define the inner product _ 1|_2=_0 .
the use of such an inner product , which employs the vacuum state as a reference vector , was suggested already in @xcite and @xcite .
as we are going to show , it leads to a discrete mesure on the space of locally flat connections , which in turn will make the proof of cylindrical - consistency of the inner product rather simple to obtain .
it is interesting to notice that other proposals using a non - discrete measure on the space of flat connections have appeared earlier in the literature @xcite , but so far do not allow for a clear understanding of the cylindrical - consistency properties of the inner product ( as opposed to the present framework ) @xcite . to simplify the discussion ,
let us now consider that the configuration space is given by a single copy of the ( compact and semi - simple ) gauge group @xmath1 , and choose the auxiliary inner product to be the one induced by the haar measure on @xmath1 .
furthermore , let us use a heat kernel regularization in order to regularize delta - peaked states . for @xmath116
, we can then define @xmath112-regulated families of states converging to the delta function peaked on @xmath117 as _
d_(-())_(g^-1 ) . here
we have chosen representatives @xmath118 , given by representation matrices @xmath119 and with dimension @xmath120 , for the equivalence classes of unitary irreducible representations of @xmath1 .
@xmath121 denotes the quadratic casimir ( the eigenvalue of the laplacian on the group manifold ) associated to the representation @xmath118 , and @xmath122 is the character of @xmath118 . since the representation matrices satisfy @xmath123 , we can readily compute = = . using the asymptotic expansion of the heat kernel on compact lie groups
( see for instance @xcite ) , which takes the form ^_(g ) ( - ) , where @xmath124 is the geodesic distance in @xmath1 between @xmath125 and the unit element @xmath126 , we finally get the following limit of vanishing regulator for the inner product : [ limitip ] ( , ) , here @xmath127 if and only if @xmath128 and is vanishing otherwise .
this delta symbol should therefore be treated as a kronecker delta with continuous parameters .
this is similar to what happens in the case of the bohr compactification of abelian groups , in particular for example in loop quantum cosmology ( lqc ) @xcite .
however , in lqc the bohr compactification typically corresponds to the configuration space ( i.e. to the holonomies , which are proportional to the momentum @xmath129 conjugate to the scale factor @xmath130 ) , whereas here , in the case of the abelian group @xmath131 , we are compactifying the dual group @xmath132 ( i.e. the space of fluxes ) , which in turn leads to a discrete topology on @xmath131 itself .
let us now turn to the basis states for the hilbert space @xmath95 . introducing the collection @xmath133 of group elements , the basis states @xmath134 are of the form [ basisvectors ] _
\{_}\{g _ } = _ ( _ , g _ ) , and are orthonormal in the above - introduced inner product : [ eq : hsinnerproduct ] _ \{_}|_\{_}= _
( _ , _ ) .
note that the hilbert space @xmath95 , although being associated to a finite triangulation , is therefore non - separable .
more precisely , the hilbert space is the closure of states of the form \{g _ } = _ \{_}f\{_}_\{_}\{g _ } , where the sum runs over a _
finite _ set of values @xmath135 and @xmath136 is a function on this finite set .
although we are here developing a flux formulation of lqg , it is worth noticing that we work still in the holonomy representation for the arguments of the wave functions . in appendix [
appendix : spin ] , we introduce a dual spin representation based on a suitable discrete generalization of the ( group ) fourier transform .
this spin representation can also be used to express the measure functional , as explained in appendix [ appendix : spin ] .
the construction which we have carried out so far depends on a specific choice of maximal tree @xmath49 with ( fixed ) root node @xmath42 .
however , it is not difficult to see that the inner product is independent of the choice of tree , which means that the construction of @xmath95 is also independent of this choice . in order to see this explicitly ,
let us choose another tree @xmath137 in @xmath34 with the same root @xmath42 .
the trees @xmath49 and @xmath137 define two different isomorphisms between @xmath94 and @xmath78 , which means that the same connection in @xmath94 corresponds to two different sets @xmath138 and @xmath139 of group elements .
the explicit isomorphism between the two sets of group elements can be constructed as follows . for both sets @xmath53 and @xmath140 of leaves , the corresponding sets @xmath141 and @xmath142 of cycles are fundamental .
therefore , a given cycle @xmath54 can be expressed as a word in the cycles @xmath142 ( and their inverses ) , and the other way around , where the orientation ( and ordering ) of @xmath54 in the product is according to the orientation ( and ordering ) in which the leaves @xmath51 appear in @xmath143 . ] .
we denote these relations by _
= w_\ { _ } , _
= w_\{_}. because of this , the corresponding group holonomies for the same flat connection in @xmath94 can therefore be expressed as [ eq : isomorphism ] g_=w_\{g_},g _
= w_\{g_}. let us now consider a state @xmath144 in the representation based on the choice of tree @xmath49 .
a change of tree @xmath145 induces a gauge transformation @xmath146 of the function @xmath147 on @xmath78 , which we denote by @xmath148 . the gauge - transformed state is then given by \{g _ } = ( ) \{g _ } = \{w_\{g_}}. in particular , for the basis states @xmath134 we have that ( _ \{_})\{g _ } = _ \{_}\{w_\ { g _ } } = _ ( _ , w_\{g _ } ) = _ ( w_\{_},g _ )
= _ \{w_\{_}}\{g_}. here we have used the fact that the words @xmath149 ( or their inverses @xmath150 ) define a bijective map @xmath151 , which means that @xmath152 if and only if @xmath153 . with these ingredients , we can finally compute the inner product using the representation based on the choice of tree @xmath137 , which leads to ( _ \{_})|(_\{_})=_\{w_\{_}}|_\{w_\{_}}= _ ( _ , _ ) .
this is identical to the inner product between states expressed with respect to the choice of tree @xmath49 , which shows that the inner product is independent from the choice of tree , or in other words invariant under gauge transformations not acting at the root .
we are now going to deal with these residual gauge transformations . the states in @xmath154 which we have constructed correspond to states on almost gauge - invariant connections @xmath94 .
under the gauge transformation at the root @xmath42 , the group elements in @xmath155 become conjugated as [ eq : gaugetransformationatroot ] \{_}^u = \{_1
,_||}^u \{u_1u^-1, ,u_||u^-1 } , where the group element parametrizing the gauge transformation is @xmath156 .
due to the form of the inner product , it is clear that the gauge - averaged states which result from integrating the action over the gauge group @xmath1 have infinite norm . therefore , the hilbert space @xmath157 of gauge - invariant states has to be constructed differently , and one way to do so is by using the technique of refined algebraic quantization ( raq ) @xcite . for this ,
let us denote by @xmath158 the dense subspace of finite linear combinations of @xmath134 with in @xmath34 , it is easy to see that @xmath158 is actually invariant under a change of tree . ] @xmath135 . instead of being the image of a projector on @xmath154
, the hilbert space @xmath157 of fully gauge - invariant functions is going to be obtained as a subspace of the algebraic dual @xmath159 of @xmath158 .
these various spaces are organized in a so - called gelfand triple _ _ _ , where the last inclusion is given by the riesz representation theorem , using the inner product on @xmath95 .
the elements in @xmath159 can be thought of as distributional states . instead of a projector onto gauge - invariant states , raq employs a _ rigging map _ ,
i.e. a linear function @xmath160 which maps to the gauge - invariant distributional states in the sense that _ = _ for all @xmath161 .
we choose this rigging map to be [ eq : riggingmap ] _
_ g/__(u)_\{_}^u|=_ug_\{_}^u| , where @xmath162 is the stabilizer of @xmath163 under the adjoint action of @xmath1 , and where we use the discrete measure @xmath164 on @xmath1 .
because an element @xmath165 is a finite linear combination of basis elements @xmath134 , the integral in is finite since only finitely many elements in the sum are non - vanishing . on the image of @xmath166
, we can now define the pre - inner product form [ eq : preinnerproduct ] _
|___\ { _ } = \ { l , + . since is sesquilinear and non - negative , the image of non - zero vectors which have vanishing norm in @xmath166 can be completed with respect to it and we can construct the hilbert space as the closure ^u_. because the discrete measure @xmath164 is translation - invariant on @xmath1 , the kernel of @xmath166 is generated by elements of the form @xmath167 . in other words , gauge - equivalent states are mapped to the same distributional state . as a result , an orthonormal basis of states in @xmath157
is given by states which are labelled by gauge - orbits @xmath168 $ ] in @xmath78 .
the inner product in @xmath157 is such that two such states have zero norm precisely if two of these orbits do not coincide , i.e. _ [ \{_}]|_[\{_}]=([\{_}],[\ { _ } ] ) .
it should be noted that the isomorphisms commute with the gauge transformations at the root , which implies that the resulting hilbert space @xmath157 is of course independent of the choice of tree .
finally , note that there exist in principle alternative versions of the rigging map which include an @xmath168$]-dependent factor . with these other choices ,
the resulting inner product will simply change by state normalization .
now that we have introduced the hilbert space , the inner product , and a basis of states on a fixed triangulation , we are ready to discuss the quantization of the algebra of holonomy - flux observables .
this will then enable us to discuss , in the next section , the construction of the area operator and the associated notion of quantum geometry .
in the previous section , we have focused on the configuration space of connections associated to a triangulation . in order to obtain the phase space of lqg , we need to supplement the holonomies encoding the connection degrees of freedom with their conjugated fluxes of the electric field .
this will allow us to discuss in which sense the quantum operators which we are going to define provide a representation of the holonomy - flux algebra . as discussed in @xcite , the finite - dimensional ( almost ) gauge - invariant phase space @xmath169 associated to a triangulation @xmath33
is spanned by cycle holonomies @xmath77 and rooted fluxes @xmath170 associated to the leaves .
these variables , which form the rooted holonomy - flux algebra @xmath171 , are defined as @xmath172 and @xmath173 , where @xmath174 is the holonomy along the tree going from the root to the source node of the leaf @xmath51 .
the flux @xmath175 is an element of the lie algebra of @xmath1 . in the case
where @xmath176 for example , the poisson brackets between these variables are given by ^i_,^j_=_,^ij_k^k_,^k_,g_=_,g_^k-_^-1,^kg_,g_,g_=0 , where the @xmath177 denote the generators of @xmath178 .
this is nothing but the poisson structure on @xmath179 , where @xmath180 .
note that the same poisson algebra ( however with dirac brackets ) , between the holonomies @xmath181 and fluxes @xmath175 associated to the leaves , can be obtained by starting from the gauge - covariant phase space associated to the graph @xmath34 and performing a gauge - fixing holonomies associated to paths along several links , and by @xmath182 holonomies associated to single links .
therefore , the gauge - fixed data is given by elements @xmath183 , while at the almost gauge - invariant level we have @xmath184 . ] .
the gauge - covariant phase space is parametrized by holonomies @xmath81 and fluxes @xmath185 associated to all the links of the graph , and the poisson structure is that of @xmath186 . by choosing a tree , one can gauge - fix all the branch holonomies to be @xmath187 .
the associated dirac brackets , involving @xmath181 and @xmath175 only , do then agree with the poisson brackets and lead again to the poisson structure on @xmath179 . since we are working in the connection representation , we can expect that the holonomies will be quantized as multiplication operators and the conjugated fluxes as derivative operators .
while it is true that the holonomies will act by multiplication , the fluxes can however not be realized as derivative operators .
this is due to the peculiar hilbert space topology which we have chosen in order to accommodate the flat state as a normalizable state an to allow for the construction of an inductive limit hilbert space .
states in this hilbert space , seen as functions on @xmath78 , are generically not continuous .
this effect is analogous to what happens in lqc due to the bohr compactification of the configuration space ( remember however , from the discussion of section [ overview ] , the difference between lqc and the construction based on the bf vacuum presented here ) . the way around this difficulty is to introduce exponentiated derivative operators , or , in other words , translations .
indeed , we are going to see that the quantized exponentiated fluxes act as left or right translations on the appropriate copy of the group .
in addition to considering the exponentiated version of an elementary flux variable , we will also have to discuss the action of exponentiated parallel - transported fluxes .
it turns out that the parallel transport will appear as an adjoint action of certain holonomies on the group element parametrizing the translation . to be more precise
, the exponentiation of the fluxes is to be understood as exponentiating the symplectic flow generated by the fluxes . indeed ,
if one sees the fluxes as representing vector fields ( acting as derivatives ) on the phase space ( see for instance @xcite ) , the exponentiated fluxes correspond to integrating these vector fields to a finite flow .
let us illustrate this for one copy of the group .
the symplectic flow induced by an elementary flux @xmath185 associated to a link @xmath82 is given by [ flux1 ] ( _ ix^i_l,)f(h_l ) _
k\{_ix^i_l , f(h_l)}_k = r_l^(_i^i)f(h_l ) f(h_l(_i^i ) ) , where the iterated poisson brackets are defined by @xmath188 and @xmath189 . for the link with a reversed orientation , we have that ( _ ix^i_l^-1,)f(h_l ) r_l^-1^(_i^i)f(h_l ) = l_l^(_i^i)f(h_l ) = f((-_i^i)h_l ) . in an abuse of notation
, we will denote the group element parametrizing this flow by @xmath190 .
so far we have considered the flow induced by an elementary flux which is not parallel - transported .
note that if we understand this description as arising from a gauge - fixing , then the flux @xmath175 in the gauge - fixed version does indeed correspond to the rooted ( parallel - transported along the tree ) flux @xmath170 .
hence , the ( exponentiated ) action of @xmath175 on @xmath181 agrees with the ( exponentiated ) action of @xmath170 on the cycle holonomy @xmath77
. we might however encounter further parallel transport for the fluxes , not necessary along the tree . in the almost gauge - invariant description
, such a parallel transport will be along a closed loop @xmath48 , and will appear through the adjoint action of a product @xmath191 of cycle holonomies @xmath192 , which in the gauge - fixed description is just given by the corresponding product @xmath193 of leaf holonomies @xmath194 .
let us assume for now that @xmath181 is not appearing in this product .
we then have for the symplectic flow generated by @xmath195 the following action : ( _ i\{h^-1_x^i_h_,})f(h _ ) = f(h_h_h_^-1 ) . if @xmath193 includes a leaf holonomy @xmath181 ( such that it can not be reduced by using the inversion formula @xmath196 ) the evaluation of the flow does become very involved ( since the flux @xmath175 is meeting in the iterated poisson brackets the element @xmath181 in the parallel transport ) .
we will exclude this case and not allow such a parallel transports to happen . in summary , we are going to quantize the parallel - transported fluxes @xmath197 in their exponentiated form as parallel - transported ( right ) translations , i.e. g_^-1_g _
r_^g_g_^-1 , and the parallel - transported fluxes associated to inverted links as left translations , i.e. g_^-1_^-1 g _ l_^g_g_^-1 . in all cases ,
we assume that the parallel transport @xmath191 does not contain the holonomy @xmath181 ( except in the case that the right translation can be rewritten into a left one and vice versa ) .
note that although a representation of the fluxes as derivative operators is not available , it is still possible to approximate derivative operators in terms of translations generated by the exponentiated fluxes .
this is what we will use in the next section in order to discuss the area operator .
we have here considered only the fluxes associated to the leaves .
it is however possible to reconstruct the fluxes associated to the branches starting from the ones associated to the leaves by solving the gau constraints .
the corresponding translation operators are then products of translation operators associated to the leaves .
this construction is explained in appendix [ appendix : branchx ] .
let us consider the space @xmath198 of continuous functions @xmath199 over @xmath79 copies of the gauge group .
being continuous functions on a compact space , these functions are bounded , and in the holonomy representation the holonomy operators acting as multiplication of states by elements of @xmath198 are therefore bounded operators . examples of such functions are matrix elements @xmath200 for a unitary irreducible representation @xmath118 .
although states of the form @xmath201 are not normalizable in our inner product ( if @xmath1 is not a finite group ) , such functions are allowed to be used as multiplication operators , i.e. lead to densely defined operators .
another example is given by functions of the form f\{g _ } = _ if_i_(g_,^(i ) _ ) , where the index @xmath202 runs over a finite set , and @xmath203 is the @xmath51-th entry in the @xmath202-th element @xmath204 of a finite subset of @xmath78 .
again , we denote by @xmath205 the kronecker delta as defined below equation . the action of an holonomy operator on a state in @xmath95 is given by [ holop1 ] \{g _ } = f\{g_}\{g_}. the holonomy operators are straightforward to diagonalize , and the eigenvectors agree with the basis vectors @xmath134 .
therefore , such an operator has a discrete spectrum given by the image of @xmath206 in @xmath207 ( or in @xmath9 if one takes the real or imaginary parts of @xmath206 ) .
note that the spectrum can appear as continuous , but always comes equipped with a discrete spectral measure .
as discussed in the beginning of this section , we do not quantize the fluxes themselves , but rather their exponentiated version , which leads to an action as right translation operators . to the flux associated to the @xmath202-th leaf @xmath208 we associate the operator r__i^\{g _ } = ( g_1, ,g_i, ,g_|| ) .
a ( first order ) difference operator can then be defined as d^__i ( r__i^-r__i^^-1 ) , with @xmath209 denoting the geodesic distance in @xmath1 between @xmath210 and @xmath126 .
the translation operators are isometric since the discrete measure employed for the inner product on @xmath95 is invariant under translations .
they are also invertible , i.e. ( r_^)^-1 = ( r_^)^=r_^^-1 , and therefore unitary .
it is quite instructive to discuss the spectrum of such a translation operator , and to compare it to the spectrum of a flux operator in the al representation
. this will be the focus of the next subsection .
let us now also discuss the action of parallel - transported exponentiated fluxes , which , as discussed above , are quantized as operators acting in the following way : r__i^g_g_^-1\{g _ } = ( g_1, ,g_ig_g_^-1, ,g_|| ) .
these operators are also isometric ( and unitary ) because of the invariance of the discrete measure under translations .
the action of a parallel - transported exponentiated flux on a basis state is given by r__i^g_g_^-1_\{_}\{g _ } & = _
\{_}(g_1, ,g_ig_g_^-1,
,g_|| ) + & = ( _ i_^-1_,g_i)_i(_,g _ ) + & = _ \{\{_}_i;_i_^-1_}\{g _ } , where @xmath211 denotes the same product in terms of @xmath163 as @xmath191 in terms of the fundamental cycle holonomies @xmath77 ( remember that we do not allow the link @xmath208 to appear in the parallel transport , as otherwise the last equality would not hold ) .
we are now going to discuss the spectrum of the translation operator . for simplicity ,
let us focus on one single copy of the group , and consider the hilbert space @xmath212 , where @xmath164 denotes the discrete measure .
this hilbert space being non - separable , the spectral analysis is a priori not straightforward to carry out .
fortunately , since the translation operator on @xmath2 and @xmath131 is invertible , it leaves separable subspaces of @xmath213 invariant , and its spectral analysis can therefore be done on each separable subspace separately . for a fixed @xmath116 , consider the invariant subspace _
^= , which is the closure of the space of states obtained by applying an arbitrary integer power of the translation operator to a basis state @xmath214 .
we can transform the matrix representation of the group with the adjoint transformations @xmath215 and @xmath216 , in such a way that @xmath217 is diagonal with @xmath218 , and restrict ourselves to considering the translation operator for @xmath131 .
let us therefore turn to the spectral analysis of the translation operator @xmath219 on @xmath131 equipped with the discrete topology .
we consider basis functions @xmath220 given by @xmath221 , which we declare to be orthonormal in the kronecker delta inner product @xmath222 , where the arguments are understood modulo @xmath223 .
these functions form a basis of the non - separable hilbert space = , a typical element of which is given by a function of the form ( ) = _ if(_i)__i ( ) , where @xmath202 takes values in some finite subset of @xmath224 , and @xmath225 is the @xmath131 angle . the inner product between two such functions is simply given by _ 1|_2=_()_2 ( ) = _ if_2(_i ) .
the action of the translation operator on elements of @xmath213 is given by r^ ( ) = ( + ) , and its action on basis functions by .
on basis vectors @xmath134 , the group elements @xmath226 evaluate to @xmath227 ( where @xmath228 does not appear ) .
the discussion is therefore exactly the same , since the class angle of @xmath210 is invariant under conjugation . ]
r^_=_- . as mentioned above , since this operator is invertible it leaves separable subspaces of @xmath213 invariant .
the spectral analysis can therefore be done separately on each separable subspace .
let us consider the sub - hilbert space which is the closure of the space of states generated by applying an arbitrary integer power of the translation operator to the basis @xmath214 , i.e. _ ^ , where the label @xmath117 in @xmath214 is to be understood in @xmath9 modulo @xmath223 .
the properties of this subspace and of the translation operator acting on it will depend on whether the angle @xmath229 parametrizing the translation is rational or not .
we now discuss these two cases separately . 1
. + this case corresponds to @xmath230 , which means that @xmath231 with @xmath232 such that @xmath233 .
then we have that @xmath234 , and the repeated action of @xmath219 on a basis state @xmath214 generates finite - dimensional ( in fact @xmath235-dimensional ) subspaces _ ^\{_-n|}. in terms of these finite - dimensional spaces , the whole hilbert space decomposes as = _ [ 0,2/q)^_. it is straightforward to diagonalize the translation operator on the finite - dimensional spaces @xmath236 .
in particular , the normalized eigenvectors are of the form [ eigen1 ] v _ , = ^q-1_n=0e^in_+n , and the associated eigenvalues are ( r^ ) = \{e^i|=0, ,q-1}. let us now fourier transform these eigenvectors using the ( group ) fourier transform introduced in appendix [ appendix : spin ] ( this will also help illustrate the case of an irrational angle @xmath229 ) . in the case of @xmath131 , the momentum representation is defined with the discrete measure by ( k ) = _ ( ) e^-ik ( ) , where @xmath237 . the inner product expressed in @xmath44-space ( and which actually defines the bohr compactification on @xmath132 ) is [ inp ] _ 1|_2=_t_|k|t_2(k ) .
one would expect the functions @xmath238 to be eigenvectors of the translation operator . however , these functions have vanishing norm in the inner product .
let us therefore consider the @xmath44-representation of the eigenvectors . with @xmath239
, we have v_,(k ) = _ n=0^q-1e^in_+n(k ) = e^-ik_n=0^q-1e^in(-k ) = e^-ik_(q)(,k ) , where we have used @xmath231 , the summation representation of the periodic kronecker delta @xmath240 ( with arguments modulo @xmath235 ) , and the fact that @xmath241 since @xmath235 does not divide @xmath242 .
therefore , instead of vectors @xmath243 , which have vanishing norm , we have eigenvectors @xmath244 with finite norm . letting
@xmath245 illustrates the fact that we have to expect generalized eigenvectors in the case of an irrational angle .
this case corresponds to @xmath246 .
the angle @xmath117 then labels an equivalence class @xmath247 $ ] of angles defined by the equivalence relation [ eclass ] .
let us introduce the set @xmath248 of all such equivalence classes , as well as the ( countably ) infinite - dimensional hilbert spaces _
^. the whole hilbert space @xmath249 then decomposes as = _ [ ] i_^. we prove in appendix [ appendix : spectrum ] that the spectrum of the translation operator @xmath219 on each subspace @xmath236 consists of the whole circle , i.e. ( r^ ) = ( 1 ) , and is continuous ( in the sense that the spectral measure is continuous ) .
the generalized eigenvectors are given by [ irreigen ] w _ , = _
ne^in_+n , with generalized eigenvalues @xmath250 where @xmath251 . choosing a different representative @xmath252 $ ] in the equivalence class @xmath247 $ ] defined in just leads to a phase factor for @xmath253 . in @xmath44-space , where again @xmath239 , the generalized eigenvectors are given by w_,(k ) = _ ne^in_+n(k ) = e^-ik_ne^in(-k ) = 2e^-ik(-k ) , where @xmath254 is now the delta _ function _ with respect to the continuous measure @xmath255 on @xmath131 .
we can in fact understand these generalized eigenvectors via a regularization procedure which uses the variable @xmath256 in the inner product as a cutoff .
one considers the inner product with fixed @xmath256 and eigenvectors [ cutoffvec ] w^t_,(k ) = c(t)e^-ik(,k ) , where @xmath257 $ ] , and @xmath258 is an appropriate normalization factor . to keep the norm of @xmath259 finite when @xmath260 , the normalization factor has to grow to infinity .
let us finish this subsection by comparing the spectra of the translation operator for the different cases discussed above , with the spectra of the translation operator defined on the hilbert space @xmath261 .
these can be obtained by evaluating the function @xmath262 on the spectrum @xmath132 of the momentum operator @xmath263 .
we then have to discuss separately the two cases for the angle @xmath229 . 1 .
+ if @xmath231 with @xmath232 such that @xmath233 , then we have a discrete spectrum @xmath264 , and eigenvectors @xmath238 in the @xmath44-space representation which now have finite norm .
if @xmath229 is irrational , then we have a `` discrete '' spectrum @xmath265 , in the sense that the spectral measure is discrete since it derives from that of the momentum operator with discrete spectrum .
furthermore , the eigenvectors are again given by @xmath238 .
we therefore see that the difference between the two quantizations expresses itself in a rather subtle way , namely in the different nature of the spectral measure for the translation operator with irrational angle . now that we have an understanding of these subtleties at the level of the translation operator
, we can turn to the more interesting case of the area operator .
in this section we are going to introduce the area operator and discuss some of its properties .
for this , we focus on the case with @xmath16 spatial dimensions .
the length operator in @xmath15 spatial dimensions exhibits similar features , but it is more instructive to study the three - dimensional case in order to grasp all of the subtleties and to understand the role of the barbero immirzi parameter .
recall that the fluxes arise as the smearing of the densitized triad fields , which themselves encode the information about the intrinsic geometry , and in particular the spatial metric tensor .
the area of a surface can therefore naturally expressed in terms of the fluxes .
let us fix a surface @xmath266 , which we can think of as being built from minimal triangular surfaces spanned by piecewise geodesic arcs .
it is therefore naturally a triangulated surface @xmath267 , and we denote its elementary triangles by @xmath268 . the area of this triangulated surface is given by [ classical - area ] _
_ t_s|_i=1,2,3^i_t^i_t|^1/2 _ t_s_t , where the flux @xmath269 associated to a triangle @xmath270 is going to be defined below . here we have introduced the barbero
immirzi ( bi ) parameter and its conjugated electrical field variable @xmath271 .
the barbero
immirzi parameter appears in the ashtekar
barbero connection through @xmath272 , where @xmath273 is the spin connection and @xmath274 the extrinsic curvature ( contracted with a triad ) . the electric field result from a rescaling by @xmath275 of the densitized triad fields .
therefore , the geometric fluxes are given by the fluxes @xmath276 multiplied by @xmath277 . ]
@xmath277 , and set @xmath278 . before quantizing this expression , let us collect a couple of remarks which already highlight at the classical level some key differences between the area operator in the bf representation and in the al one .
* the subdivision of the surface @xmath266 into a set @xmath267 of triangles @xmath268 is an essential ingredient of the definition .
the choice of @xmath267 provides a `` scale '' on which the area will be measured , which implies in particular that two different triangulations of @xmath266 may a priori lead to different results .
this should not be seen as a disadvantage of the framework . on the contrary
, it addresses the so - called staircase ( or `` coast of britain '' ) problem for fractal geometries , which is that since the measurement of a given length depends on a choice of measuring scale , for fractal - like geometries one can in principle get infinite results in the limit where the measuring scale approaches zero . in the present framework , the triangles @xmath268 play the role of the smallest available rulers with which one can measure the geometry . as usual in loop quantum gravity , `` smallest '' should here be understood in relative terms , since the metric properties are encoded in the states themselves .
what actually matters is rather the fineness of the triangulation @xmath267 as compared to the fineness of the states onto which the area operator is applied .
+ since one can choose @xmath267 to be much coarser than the triangulations underlying the states , the bf - based representation provides a natural way of defining coarse - grained geometric quantities .
this possibility does a priori not exist in the al representation , which leads to problems when trying to construct suitable semi - classical states approximating smooth geometries ( see for instance the discussion in @xcite ) .
+ because of its behavior under refinement operations , the area operator introduced here is similar to the one proposed by livine and terno in @xcite .
this latter is an alternative to the usual area operator in the al hilbert space @xcite .
the usual area operator @xcite involves a sum over the casimir operators for each spin network edge piercing the surface . under refinement of the states ,
the sum is extended in order to include the casimir operators associated to the additional edges .
the alternative suggestion made in @xcite is to first couple the representation spaces associated to the spin network edges going through a piece of surface ( which here is defined to be a triangle in @xmath267 ) to a total angular momentum vector space .
one should then considers the @xmath2 casimir with respect to this total angular momentum .
therefore , under refinement one still considers just one casimir operator , but now for a different total angular momentum vector space arising from more recoupling steps . + the difference with our new proposal is that here we have to replace the angular momentum operators with exponentiated operators ( due to the compactification of the lie algebra of @xmath2 ) , while the behavior under refinement of states is similar .
indeed the cylindrical - consistency of the area operator ( i.e. its consistent behavior under refinement ) in the bf - based representation is ensured by the recoupling prescription .
this follows from the discussion in section [ extflux ] , where we define the action of exponentiated flux operators on refined states , which we summarize in the next two points . * the area
@xmath279 given by can be defined on all the triangulations @xmath33 which support @xmath267 , i.e. in which the triangles @xmath268 of @xmath267 arise as the possible coarse - graining of ( unions of ) triangles @xmath280 . because of this , the fluxes @xmath269 have to be thought of as composed objects .
one possibility is to define these as `` integrated fluxes '' associated to the triangles @xmath268 following the definition of @xcite ( the triangle @xmath268 is then to be understood as a co - path @xmath281 in @xcite ) , i.e. [ xt ] _
t g^-1_rl_1(0)(_l_itg^-1_l_1(0)l_i(0)x_l_ig_l_1(0)l_i(0))g_rl_1(0 ) . here
@xmath282 labels all the links which are dual to the triangles of @xmath33 composing @xmath268 , the source node @xmath283 of the link @xmath85 serves as a root for the surface @xmath268 , and @xmath284 is the parallel transport from this source node to the source node of the link @xmath282 .
this parallel transport takes place along a surface tree ( defined in detail in @xcite ) , which is as close to the surface as possible ( and below the oriented surface ) .
this means that @xmath285 ( and therefore the area operator itself ) depends on this choice of surface tree ( for every triangulation @xmath33 on which one wishes to define @xmath285 ) .
note that in we also have to consider in general fluxes associated to links that are not necessarily leaves .
such branch fluxes can however be expressed in terms of fluxes associated to the leaves , as explained in appendix [ appendix : branchx ] .
+ one can come up with alternative definitions for @xmath285 by changing the prescription for the parallel transports .
for instance , one can extend the triangulation @xmath267 to a @xmath14-dimensional triangulation @xmath286 in such a way that each @xmath268 is a boundary triangle ( with outward orientation ) of a @xmath14-simplex in @xmath286 .
we can then allow for a parallel transport which is not necessarily as close to the surface as possible , but rather confined to be inside the coarse @xmath14-simplex in @xmath286 which bounds @xmath268 .
this definition corresponds to the notion of coarse flux observables which we will define in section [ splitting algebra ] .
the area operator will then depend on this additional data @xmath286 , and on a definition of parallel transport in each of its simplices .
* with the definition(s ) of @xmath285 discussed above , we can write down a consistent area operator which can be applied to states based on different triangulations .
consistency means that the area operator will give the same result for states connected by the refinement operations which we define in section [ sec : refinement ] .
this allows for the area operator to be defined on a continuum ( inductive limit ) hilbert space .
+ given a triangulation @xmath33 supporting @xmath267 or @xmath286 , we need to specify the parallel transport for the definition of the fluxes @xmath285 .
this will then give a full definition of the area operator @xmath279 or @xmath287 for the triangulation @xmath33 .
the area operator can also be defined for all states based on a triangulation which are coarser than @xmath33 . to this end ,
one just needs to refine the coarser states to states defined on @xmath33 ( in a manner defined in the second part of this work ) .
however , the extension to states defined on triangulations @xmath46 finer than @xmath33 is less trivial .
the basic problem is to define a notion of parallel transport which leads to consistent results for arbitrary fine triangulations .
this prescription corresponds therefore to a continuum prescription , and what one needs to specify is a method for projecting this prescription to a given triangulation @xmath46 .
we outline in section [ contop ] a prescription for parallel transport in the continuum which can be projected in a consistent manner to triangulations . using this method
, we can define the area operator as a continuum operator .
* since only the exponentiated fluxes can be quantized , we have to approximate the differential operators by difference operators .
such approximations to the gauge - invariant expressions of the form @xmath288 are however not completely gauge - invariant anymore ( this concerns only the gauge transformations at the root ) .
one could therefore introduce a further change of parallel transport for the ( exponentiation of the fluxes ) @xmath285 , so that the fluxes associated to the triangles @xmath268 are first transported to some node in @xmath286 , and only then to the root .
this makes the definition of the area operator more local in its dependence on data near or far away from the surface . by first transporting the fluxes along the surface
, we will have a more local dependence .
data off of the surface appear only via the global parallel transport from the chosen node in @xmath286 to the root .
let us now discuss the quantization of the area operators .
as mentioned above , we have to approximate the fluxes with difference operators . to this end
, we choose a parameter @xmath289 , along with a basis @xmath290 , and define @xmath291 as ^_i ( ^i ) .
we then define a quantization of the quantity @xmath292 as [ areaelement ] _
i^i_l^i_l _ i(r^^+_i_l+r^^-_i_l-2 ) this quantization covers the case in which @xmath267 coincides with a triangulated surface of the triangulation of @xmath33 . in this case , the fluxes @xmath285 appearing in the expression for the area are given by the fluxes @xmath293 .
the quantized area operator is then given by [ areaoperator ] ^__s = _ l_s|-_i(r^^+_i_l+r^^-_i_l-2)|^1/2 . before discussing the spectrum of this operator , let us gather a few remarks concerning its definition . * in section [ contop ]
, we discuss the way in which the translation operators can be defined on the continuum hilbert space .
this can be applied in order to obtain a consistent continuum definition of the area operator , as discussed above for the classical expression . *
as mentioned above , the quantized area operators are not invariant under the frame rotations at the root anymore .
we therefore discuss later on in this section strategies for defining a fully gauge - invariant version of the area operator .
* interpretation of the parameter @xmath8 : the approximation of the flux derivative operator by difference operators is similar to the approximation of the extrinsic curvature ( or the momentum conjugated to the scale factor ) by ( point ) holonomies in loop quantum cosmology ( lqc ) @xcite . in fact , in lqc the configuration space of the scale factor is also quantized with a discrete topology , which leads to an evolution equation which is discrete in @xmath8 steps . in the case of lqc
, it is necessary to adjust the step parameter @xmath8 to the value of the scale factor @xcite . in our case
, @xmath8 parametrizes a translation on the group , which encodes the values of the curvature defects .
one could attempt to introduce a dependence on the class angle of the group element which is being translated , which would lead to a curvature - dependent discretization step .
this would in turn change the properties ( and the spectrum ) of the area operator .
we will discuss such state - dependent @xmath8 parameters for the fully gauge - invariant area operator . * the barbero
immirzi parameter : in , we have introduced the bi parameter by treating it as a factor which multiplies the difference operator .
there exists however an alternative ( quantization ) ambiguity , which is to define the group elements by which one translates as [ altversion ] ^_i ( _ ^i ) , and to omit the prefactor @xmath277 in . in this case
( at least for @xmath131 ) the bi parameter can redefine the values of the parameter @xmath8 which lead to discrete and continuous spectra respectively .
this alternative changes also the bound on the spectrum of the area operator : whereas in the version it multiplies the bound , version has a bound which is independent of the bi parameter . also , at least in the case of @xmath131 , if the barbero immirzi parameter is included using , the spectrum of the area operator does then not depend on this parameter as long as @xmath294 is irrational .
let us consider the area operator associated to one triangle @xmath268 , and focus on a triangulation @xmath33 where this triangle @xmath268 appears and is not further refined . with this choice ,
the triangle @xmath268 is dual to a link @xmath82 , and we have that [ arqu ] ^_l = |-_i(r^^+_i_l+r^^-_i_l-2)|^1/2 .
note that the ( squared ) area operator is a linear combination of bounded operators , and therefore is itself bounded .
the bound does however grow with @xmath295 .
this boundedness results from the compactification of the dual of the lie group , and is similar to the bound on curvature which appears in lqc due to the compactification of the momentum variable conjugated to the scale factor .
this bound on the area is also similar to the quantum group deformation ( at root of unity ) , where only a finite range of representations @xmath11 are admissible . on the other hand , it is not so clear that the spectrum of the area operator is still discrete , as it is in the al representation .
let us now discuss the spectra for the two gauge groups of interest .
we start by discussing the case @xmath296 , where the issue with gauge transformations at the root does not arise .
furthermore , in this case the translation operators commute , so we can use directly the results of the spectral analysis of the translation operators themselves .
again , the properties of the area operator will depend on whether the parameter @xmath8 is rational or irrational . 1
. + this case corresponds to @xmath297 , which means that @xmath298 with @xmath232 such that @xmath233 . then we have eigenvectors of the form v__1,_1v__2,_2v__3,_3 , where @xmath299 is given by .
this leads to a discrete spectrum \{|_i(2 - 2(_i))|^1/2|(_1,_2,_3)(z_q)^3 } for the @xmath296 area operator .
this case corresponds to @xmath300 .
we can then apply the spectral theorem on the separable invariant subspaces discussed in section [ spectrans ] .
thus , we have generalized eigenvectors w__1,_1w__2,_2w__3,_3 , where @xmath301 is given by , and a @xmath302 labels an eigenvalue @xmath303 of the translation operator @xmath304 . in order to make the similarity with the discrete case more transparent
, we can define implicitly @xmath305 as the integer for which @xmath306 mod @xmath223 holds ( remember that @xmath307 can indeed be understood as being concentrated on @xmath308 in the @xmath44-representation ) . in this sense ,
the spectrum is given by the closure of the set \{|_i(2 - 2(_i))|^1/2|(_1,_2,_3)(z_q)^3}. if we choose @xmath8 as well as the @xmath309 to be small , one finds a very good approximation to the usual case , i.e. spectral values @xmath310 . for growing @xmath309 , the bound sets in and
one has an oscillatory behavior which fills densely all the allowed range , which is given by .
let us now turn to the more complicated case of the gauge group @xmath2 . here ,
the translation operators @xmath304 do not commute , so we can not apply the results from the spectral analysis of these operators straightforwardly .
however , we can define as before subspaces which are left invariant by the area operator , by taking the closure of the space of states generated by applying arbitrary integer powers of @xmath304 to a given basis state @xmath214 . if the parameters @xmath311 are elements of a finite subgroup of @xmath2 , this results in a finite - dimensional subspace with discrete spectrum .
these are however only the subgroups corresponding to the platonic solids , and thus no regular family of finite groups that generates rotations in all three directions .
therefore , we can rather expect to have a continuous spectrum , at least in the sense of having generalized eigenvectors ( it still could happen that there are normalizable eigenvectors , as explained below ) .
we can nevertheless diagonalize the area operator by employing the spin representation laid out in appendix [ spinrepsu2 ] .
this uses a ( discrete generalization of the ) group fourier transform , so that the states are wave functions of the form @xmath312 , where @xmath11 is the spin representation label and @xmath313 are magnetic indices .
the translation operators act in this representation as ( r^)(j ) = d^j ( ) ( j ) , and therefore leaves the @xmath11 label and one of the magnetic indices @xmath314 invariant .
the difference with the usual case is that @xmath315 is a state with vanishing norm , since the inner product is given by [ inpl ] _
1|_2=__j=0^((_1^(j)_2(j ) ) , with a normalization factor @xmath316 ( see ) . in order to consider the ( so far non - existant ) subspaces associated to one fixed spin @xmath11
, we introduce a fixed cutoff @xmath317 on the spins @xmath11 , and define an inner product on the space of functions @xmath318 as in but without taking the limit . the associated hilbert spaces @xmath319 are then finite dimensional ( and just correspond to ( rescalings of ) @xmath320 ) .
this enables us to consider vectors of the form [ jn ] ^jn_mn(j ) = c()a_m(j , n)_j , j_n , n , where @xmath321 is a normalization factor which grows to infinity if the regulator is taken away , i.e. in the limit @xmath322 .
the action of the square of the area operator on a subspace spanned by the vectors , for fixed @xmath323 and any fixed @xmath324 , is given by the matrix a_mn^j ( ) = -_i(d^j_mn(_i)+d^j_mn(^-1 _ i)-2_m , n ) .
we now have to diagonalize these matrices and examine the eigenvalues of the area operator on the cutoff hilbert space @xmath319 .
the general behavior of the eigenvalues is similar to the case of the gauge group @xmath296 treated above , but with an important difference : for small @xmath8 and @xmath11 , the matrix @xmath325 is nearly diagonal with almost equal eigenvalues , approximating well the casimir eigenvalues @xmath326 . for growing spin
, the approximate degeneracy of the eigenvalues is lifted , and the eigenvalues spread more and more . moreover , the eigenvalues are bounded , and an oscillatory behavior sets in , as can be seen on figure [ fig : area2 ] .
however , it is not clear whether the oscillatory behavior will cover the entire allowed interval .
axis ) of the squared area operator as a function of the spin @xmath11 ( @xmath327 axis ) , for @xmath328 , @xmath329 , and @xmath330 . ] axis ) of the squared area operator as a function of the spin @xmath11 ( @xmath327 axis ) , for @xmath331 , @xmath329 , and @xmath330 . ]
so far , we have discussed the area operator associated to one single link . under refinement , this link can be subdivided into several links , and in this case the translation operators @xmath332 have to be replaced by a product of ( possibly parallel - transported ) translation operators associated to the finer links @xmath333 .
however the essential features do not change : using the intertwiner - spin representation laid out in appendix [ intertwinerspin ] , one can couple the representation spaces associated to the finer links into a total spin @xmath11 .
the discussion then proceeds as before , with the difference that one now has higher multiplicities for the eigenvalues caused by the recoupling procedure . as before ,
the spectrum is bounded by @xmath334 .
moreover , we have only considered so far the case of a basic triangle appearing in the triangulation of the surface @xmath266 underlying the definition of the area operator for this surface .
these basic area operators will usually commute with each other . in order to obtain the spectrum of the full area operator , one needs to add up the ( square root ) of each of these contributions of the basic triangle ( squared ) area operators .
the bound on the spectrum , which is given by @xmath334 , is therefore multiplied by the number of triangles appearing in the surface @xmath266 . although the basic area operators are bounded , we can achieve a large area by subdividing a given area in many pieces .
this exemplifies even more the coast of britain " paradox : the measurement scale , which here is the subdivision of the surface into triangles , determines the maximal area of this surface . to summarize our discussion on the spectrum of the area operator ,
let us emphasize that the compactification of the flux space , i.e. of the dual to the lie group , naturally leads to a compactification of the area spectrum for the area operator associated to a basic triangle .
note that the subdivision of the surface @xmath266 is inherent to the definition of the are operator associated to this ( triangulated ) surface , and not to the triangulation on which the states are defined .
therefore , by adding up the area operators associated to many triangles , we can increase the bound on the spectrum of the composed operator .
furthermore , we have seen that it is possible to account for the barbero
immirzi parameter @xmath277 in two different ways , which does also influence the bound on the spectrum .
since the spectrum associated to a given basic area operator is bounded , it will wind around in the compact range of values specified by the bound @xmath335 $ ] . for @xmath131 , with a rational angle @xmath8 this winding happens in a periodic way and results in a discrete spectrum , while for @xmath8 irrational the winding is ergodic and gives a continuous spectrum . for @xmath2
, we can also diagonalize the area operator in the spin representation .
these are indeed proper eigenvalues if one introduces a cutoff hilbert space which only admits spins smaller than or equal to @xmath317 .
however , it is not clear how the spectrum changes if the cutoff is removed .
the spectrum associated to small spins @xmath11 ( and sufficiently small @xmath8 ) approximates the usual area spectrum @xmath336 quite well . for larger spin values @xmath11 ( or larger @xmath8 ) ,
an oscillatory behavior appears .
one can expect that this leads to a dense filling of at least a part of the space of allowed values .
it will therefore be interesting to see what happens for the fully gauge - invariant area operator , which we now discuss .
the lifting of the degeneracy of the eigenvalues in figure [ fig : area1 ] and figure [ fig : area2 ] is due to the non - invariance of the area operator under frame rotations ( i.e. gauge transformations ) at the root .
let us therefore discuss possibilities for defining a fully gauge - invariant area operator .
one possible starting point is to take the action of the area operator in the spin representation , and to consider the following ( normalized ) trace of the area operator : [ areag ] a^j _ ( ) = _ ma_mm^j ( ) = .
if would give a well - defined operator ( on either the kinematical hilbert space or on the hilbert space of fully gauge - invariant wave functions ) , and also if we took the limit @xmath337 , we could read off the spectrum from the representation in . there
, the oscillatory behavior of the sine function is suppressed by a factor of @xmath338 , which leads to a discrete spectrum for sufficiently small spins @xmath11 . to answer the question whether can give an operator on the kinematical hilbert space @xmath95
, we can consider the ( kernel of the ) operator in the holonomy representation , namely ( a_^)(h ) = _
( h)a^_(h,h)(h ) , with [ akernel ] a^_(h,h ) & = _ _ jd_j_j(hh^-1)a^j _ ( ) + & = ( h,h)-__j_j((hh^-1))_j ( ( ) ) , where @xmath339 , and @xmath340 $ ] is the class angle of the group element @xmath117 .
the sum in the second term gives the ( dirac ) delta function ( without the factor of @xmath341 ) on the equivalence classes of the adjoint action , i.e. [ suml ] _
j/2_j()_j ( ) = _ ( g)(g^-1g^-1 ) , and one could in principle interpret the area operator as a translation operator acting on these equivalence classes .
however , the sum with a cutoff @xmath342 scales rather as _
j_j ( ) _ j ( ) ~\ { cl & , + ^3 & .
because of this , the kernel would be equal to the kronecker delta for @xmath343 ( since the second term is suppressed in the limit @xmath322 ) , but would be vanishing for @xmath344 .
the expression does therefore not define a useful area operator on the kinematical hilbert space .
nevertheless , it could be possible to define matrix elements in the inner product of the fully gauge - invariant hilbert space , as defined in section [ sec : raq ] , via the raq procedure .
this would lead to an averaging over the orbits of the adjoint action ( acting on the entry @xmath345 in ) with respect to the discrete measure on the group .
the difficulty is however to exchange the sum and the limit in @xmath342 in with this discrete measure integral in a well - defined way . in summary ,
the derivation of a fully gauge - invariant area operator from the kinematical version is rather involved ( this difficulty is however lifted if one changes to a quantum group at root of unity , as will be discussed in section [ sec : discussion ] ) . an alternative way to proceed
is offered by the interpretation of the kernel with respect to the continuous topology : it would then define a translation operator for the class angle .
indeed , the usual area operator is given as a laplace operator on the group if we work with @xmath346 .
we could therefore consider the laplace operator restricted to class functions , which is [ lap ] = _
( ^2 _ ) , and approximate this differential operator by a difference operator .
this ( discretization ) procedure does however involve lots of ambiguities , and one would rather wish for a more natural connection with the area operator defined on the kinematical hilbert space .
let us therefore study how the translation operators @xmath347 , which are instrumental for the area operator on the kinematical hilbert space , act on the class angle of the argument @xmath125 being translated in the wave function .
for this , let us use the parametrization g=(g)-i(g)(g),_i^=i_i , where @xmath348 are the pauli sigma matrices , @xmath349 is a unit vector , and @xmath350 . with this
, the class angle of the product @xmath351 is given by [ classa ] ( g_i^ ) = ( ( g)((g))_i(g ) ) .
the problem is now that this class angle depends through @xmath349 on the full group element @xmath125 , and not only on its class angle .
to get rid of this dependency , we can integrate over @xmath161 the adjoint action @xmath352 , which happens to be equal to a rotation of the vector @xmath349 over the sphere @xmath353 .
this will lead to the same averaged class angles for the product of @xmath125 with @xmath354 and @xmath355 , and for all @xmath356 . with the parametrization n_1=,n_2=,n_3= , where @xmath218 and @xmath357 $ ] , the integration is easiest to perform for @xmath358 . choosing the normalized invariant measure on the sphere ,
we obtain r^((g ) ) & _ 0 ^ 2_0 ^ 2(g_2^+ ) + & = _
-1 ^ 1((g)-x(g))x + & = ( -((g)+)+ + & + ( ( g)-)-).[theta1 ] here one can choose the roots such that @xmath359 , and furthermore ( ( x ) ) = \ { cl -x & + x & + -x & .
this ensures that the result is finite for @xmath360 ( and equal to @xmath361 ) and for @xmath362 ( and equal to @xmath363 ) , as can be seen on figure [ fig : theta1 ] . between the translated and untranslated class angles , with translation parameter @xmath364 , as a function of the untranslated class angle @xmath365 .
] consider now a fully gauge - invariant hilbert space @xmath157 as defined in section [ sec : raq ] , with a basis of wave functions @xmath366}$ ] , and an area operator acting on a leaf @xmath51 .
we parametrize the orbits @xmath168 $ ] under the adjoint action in such a way that the leaf @xmath51 on which the area operator is acting appears only with a class angle , i.e. @xmath168=\{\theta_\ell,\dots\}$ ] .
we can then define the square of the area operator as [ areainv1 ] _ ^_\ { _ , } = -(_\{r^(_), }-_\ { _ , } ) .
we therefore have a translation operator acting on the class angle with a configuration - dependent translation parameter .
this makes the investigation of the spectrum more complicated , and we leave this rather important question for future work .
it should be emphasized that is one particular proposal for a fully gauge - invariant area operator , obtained by averaging the shift in the class angle over the orbits of the adjoint action .
it is certainly possibly to define alternative versions , for instance by discretizing the action of the laplacian on class functions .
we now turn to the second main part of this work , which concerns the construction of the continuum inductive limit hilbert space . for this
, we are going to introduce in section [ sec : refinement ] a few technical results concerning the refinement operations .
this will then enable us to establish the cylindrical - consistency of the inner product , which is required in order for the inductive limit to be constructed .
we will then discuss the relationship between continuous operators and operators defined on a fixed triangulation .
in this section , we are going to construct refining maps which embed hilbert spaces @xmath95 into hilbert spaces @xmath108 associated to finer triangulations .
these refinement maps are needed in order to obtain the continuum hilbert space @xmath367 as an inductive limit from the family of hilbert spaces @xmath95 . since our description of the hilbert spaces @xmath95 involves choices of trees ( for describing bases for these hilbert spaces )
, we are first going to discuss the notion of refined trees .
this will then allow us to express the refinement maps in a compact way .
it will also facilitate the expression of data on a finer phase space in terms of data on a coarser phase space , which will be used later on in order to provide extensions of the holonomy - flux observables associated to a given triangulation to a finer triangulation . in order to help the reader
, we provide here a short summary of the various notations which are used throughout the rest of this work to describe structures related to graphs , trees , and their refinements .
1 . ll & + ^-1(l ) & + ^-1_(l ) & + & + & + _ & + _ & + _i & + _i & let us consider two triangulations @xmath33 and @xmath46 such that @xmath47 , along with their respective dual graphs @xmath34 and @xmath101 . in @xcite
, we have described the projection maps @xmath368 which send paths @xmath369 in @xmath101 ( which can be viewed as paths in @xmath46 ) to paths @xmath48 in @xmath34 ( viewed as paths in the coarser triangulation @xmath33 ) .
let us recall how simplices and paths in @xmath33 and @xmath46 ( and their dual graphs ) are related by @xmath368 .
since @xmath46 is a refinement of the triangulation @xmath33 , we can identify each simplex @xmath370 of @xmath33 which is involved in the refinement with a union @xmath371 of simplices ( of the same dimension as the simplices @xmath370 ) in the finer triangulation @xmath46 .
in other words , @xmath371 gets coarse - grained to @xmath370 under the inverse of the refining map .
we denote this relationship by to denote a map from the space of path in @xmath101 to the space of path in @xmath34 , and for denoting the relationship between simplices @xmath370 in @xmath33 and the set of simplices @xmath372 in @xmath46 into which the simplices @xmath370 are refined . ] @xmath373 .
consider now a path @xmath369 in @xmath101 .
when seen as a path in @xmath46 , this @xmath369 goes inside @xmath14-dimensional simplices and intersects some of their boundary @xmath31-dimensional simplices .
we define the path @xmath374 in @xmath33 in such a way that it enters and leaves a @xmath14-dimensional simplex @xmath375 in the same order and through the same boundary simplices @xmath376 as @xmath369 enters and leaves the complex @xmath377 with respect to the boundary complexes @xmath378 .
an important property which follows from this definition of the projection maps @xmath368 is their transitivity .
this can be summarized in the following result ( here we denote by @xmath379 the projection map from @xmath46 to @xmath33 ) : [ lemma : ptransitivity ] given three triangulations @xmath380 and the corresponding projection maps @xmath379 , @xmath381 , and @xmath382 , then we have that [ ptransitivity ] _ ,_, = _ , .
every link @xmath37 , whether it is a leaf or a branch of a tree , has a pre - image @xmath383 in the set of paths in @xmath101 whose nature depends on the @xmath14-dimensional simplices dual to the end nodes @xmath39 and @xmath40 . if the @xmath14-dimensional simplices dual to the end nodes of @xmath82 are not involved in the refinement of the triangulation , then the pre - image @xmath383 consists of a single link @xmath333 . if at least one of the @xmath14-dimensional simplices dual to the end nodes of @xmath82 is being refined , then the pre - image @xmath383 consists of a set of paths ( which can a priori differ in their source and/or target nodes ) and therefore contains several links . in general , @xmath383 will include paths composed out of several links of the finer triangulation .
it will turn out to be useful to isolate the paths which include only one link . for this , we therefore introduce the notation @xmath384 for the `` link pre - image '' , i.e. the subset of @xmath383 containing all the links @xmath333 such that @xmath385 . this link pre - image will be essential for discussing the refinement of flux observables , which are based on so - called co - paths ( a set of connected edges in @xmath15 and a set of connected triangles in @xmath16 ) .
basically , @xmath386 contains all the links which are dual to the @xmath31-dimensional simplices appearing in the refinement of the @xmath31-dimensional simplex dual to the link @xmath82 .
we are now going to use these projections of paths to characterize refined trees as , roughly speaking , trees whose fundamental cycles can be projected to the fundamental cycles determined by the coarser tree . consider a triangulation @xmath33 , its dual graph @xmath34 , and a finer triangulation @xmath46 obtained from @xmath33 by alexander moves .
given a tree @xmath49 in @xmath34 , it turns out that not all the trees @xmath387 will be useful in our construction , not even if they satisfy the condition @xmath388 .
this is illustrated with some examples in appendix [ appendix : trees ] . in what follows
, we will therefore characterize the trees @xmath387 which arise as `` natural extensions '' of @xmath49 , and call them refined trees .
[ def : refine - tree ] consider two triangulations such that @xmath47 , and a tree @xmath49 in the graph dual to the coarser triangulation @xmath33 .
we say that a tree @xmath389 is a refinement of a coarser tree @xmath49 if the following two conditions are satisfied : * all the paths along @xmath389 are mapped under @xmath368 to paths along @xmath49 , which we write as @xmath388 . * for every @xmath14-dimensional simplex @xmath390 being refined when going to the finer triangulation @xmath46 , the nodes dual to any two simplices in the union @xmath391 can be connected by a path in @xmath389 without ever crossing a @xmath31-dimensional simplex in the boundary @xmath378 .
from these two conditions , it follows in fact that * _ there exists a subset @xmath392 of the set of leaves of @xmath389 such that all the fundamental cycles determined by @xmath393 can be mapped bijectively to the fundamental cycles determined by @xmath53 .
_ one can select such a set @xmath393 in the following way : for a given leaf @xmath51 in @xmath53 , consider the corresponding dual simplex @xmath394 and the subdivision of this simplex under refinement , i.e. the set @xmath395 of simplices . because of condition @xmath396 , all the simplices in this set are dual to leaves with respect to the refined tree @xmath389 .
one can then choose one of the leaves @xmath397 dual to one of these simplices to be in @xmath393 and to `` represent '' the leave @xmath51 . by making this choice for all the leaves @xmath398
, we determine the set @xmath393 .
note that this choice specifies additional information on top of the refined tree , and in the following we implicitly assume that with the refined tree we have also chosen such a subset .
we also adopt the convention according to which the orientations of the leaves @xmath399 coincide with the orientations of the corresponding leaves in @xmath53 . therefore , if necessary , we need to invert the orientations of some of the leaves @xmath397 .
one can replace condition @xmath396 with @xmath400 , which results in a weaker notion of refined tree , as illustrated with the examples of appendix [ appendix : trees ] .
this weaker notion still allows for the identification of constraints in section [ constraints ] .
however , condition @xmath396 makes the splitting of ( in particular the flux ) observables , into coarse and fine ones ( detailed in section [ splitting algebra ] ) much easier .
there is a way to explicitly construct refined trees , which we will now explain .
the construction follows two steps , and requires some choices along the way . 1 . for each @xmath14-dimensional simplex @xmath401 in @xmath33 ,
consider the collection of @xmath14-dimensional simplices which build up @xmath401 , i.e. the set @xmath402 .
choose a maximal tree dual to this piece of the triangulation , i.e. a set of edges connecting the simplices @xmath403 , which meets every vertex , but which has no closed loops . by doing this for all the @xmath14-dimensional simplices in @xmath33
, one gets a ( non - necessarily maximal ) tree in @xmath101 .
2 . for each branch in @xmath49
, take the dual @xmath31-dimensional simplex @xmath404 in @xmath33 , and consider the collection of @xmath31-dimensional simplices in @xmath46 which build up this simplex , i.e. @xmath405 .
choose one of these simplices , and add its dual link to the list of branches in @xmath389 . by doing this for all the branches in @xmath49 ,
one ends up with a maximal tree @xmath389 in @xmath46 .
it is easy to check that the properties @xmath406 and @xmath396 of definition [ def : refine - tree ] are satisfied for the resulting maximal tree @xmath389 .
in fact , it can be shown that every refined tree is the result of such a construction process .
a proof of this fact is given in appendix [ app : construction ] . in the following section , we are going to use different subdivisions of the set of leaves @xmath407 of the refined tree .
we now summarize what these various leaves are .
* one subdivision of @xmath407 is into the representatives @xmath408 for the coarse leaves and the remaining leaves @xmath409 .
this corresponds to a subdivision of the holonomy degrees of freedom into coarse holonomies and finer holonomies .
the latter are constrained for configurations which arise from a refining procedure . * furthermore , for each coarse leave @xmath51 , we have a subset @xmath410 of leaves . the leaves in a given set
@xmath411 give all the links which are dual to simplices @xmath412 which compose the coarse simplex @xmath413 dual to @xmath51 .
for a leave @xmath414 we have @xmath415 . * the remaining leaves , i.e. the ones in @xmath416 , are either links `` inside '' the coarser simplices @xmath375 , or are part of @xmath417 ( which is not a subset of @xmath418 since it contains exactly one branch @xmath419 ) .
the links in @xmath411 cross a coarser @xmath31-dimensional simplex @xmath420 which is dual to a branch @xmath50 of the tree @xmath49 .
for all the leaves @xmath421 we have that @xmath422 , i.e. the corresponding cycles are mapped to the trivial cycles .
finally , let us conclude this subsection by mentioning an additional important property satisfied by the notion of refined trees , which is that of transitivity .
this can be summarized in the following lemma , which we prove in appendix [ proof : ttransitivity ] : [ lemma : ttransitivity ] let us consider three triangulations @xmath380 .
let @xmath389 be a refined tree for @xmath49 , and @xmath423 be a refined tree for @xmath389 .
then we have that @xmath423 is also a refined tree for @xmath49 .
now that we have obtained criteria for characterizing refined trees and shown that such trees can always be constructed , we are going to use this to derive the constraints satisfied by the finer cycles of the refined graph .
these constraints will in turn enable us to define the refinement maps , and then to separate the holonomy - flux algebra on the finer phase space into two mutually - commuting subalgebras . with our definition of refined trees , we had to choose a subset @xmath393 of `` coarse '' leaves in @xmath101 which are representatives for the leaves in @xmath34 .
this means that if we index the leaves of @xmath53 with small latin letters @xmath424 , then for each @xmath425 there exists a unique leaf @xmath399 such that [ pell ] _
i = ( _ ) .
we will denote this unique leave @xmath399 which maps under @xmath368 to @xmath208 as @xmath426 .
therefore , @xmath427 restricted to @xmath393 is a bijection .
we will use the symbol @xmath428 for the corresponding map , so that @xmath429 . for the remaining leaves in @xmath430 , there can
_ a priori _ be three possibilities .
the corresponding cycles can be mapped under @xmath368 to @xmath431 the trivial cycle of @xmath34 ( i.e. to the root ) , @xmath432 to a fundamental cycle @xmath433 or @xmath434 to the inverse @xmath435 of a fundamental cycle , or @xmath436 to an arbitrary non - fundamental cycle ( which can in turn be described as a combination of fundamental cycles ) .
explicitly , if we index the leaves of @xmath430 with capital latin letters @xmath437 , for each @xmath438 we can therefore define , where @xmath42 and @xmath439 stand for the trivial cycles ( i.e. the roots ) in @xmath34 and @xmath101 respectivly . ]
[ well ] w_i\{__i } ^-1_((__i ) ) as a word in the set @xmath440 of fundamental cycles determined by the leaves @xmath426 .
however , one can show that with the definition [ def : refine - tree ] of a refined tree option @xmath436 does not occur .
this is summarized in the following result : [ lemma : words ] a refined tree , together with a separation of its leaves into leaves @xmath441 and @xmath438 , is such that the words @xmath442 are either trivial or one - letter words in the cycles @xmath443 determined by the leaves in @xmath393 ( or their inverses ) .
this result is tight to condition @xmath396 in the definition of the refined trees . without this condition ,
one can obtain in words of several letters in the cycles @xmath443 .
the proof of this lemma follows from the discussion in the previous subsection , about the partition of the refined leaves into different subsets .
if the leaf @xmath444 leads to the case @xmath431 , we have @xmath445 . in this case , the leave @xmath444 will not appear in any of the sets @xmath446 or @xmath447 . such a leave @xmath444 does not cross any of the @xmath60-dimensional simplices in @xmath46 which coarse grain to the @xmath60-dimensional simplices in @xmath33 . if the leave @xmath444 leads to the cases @xmath432 or @xmath434 , then there exists a unique @xmath208 such that either @xmath432 @xmath448 or @xmath434 @xmath449 .
accordingly , we have @xmath450 or @xmath451 .
let us now discuss how to obtain the constraints on the finer cycles of the refined tree . from equation
, it follows that we have the composition of paths @xmath452 , where @xmath42 is the trivial path ( i.e. the root ) . in terms of holonomies
@xmath453 along the paths @xmath454 , this means that we have found a set of independent constraints given by [ constraints1 ] _ iw_i\{g__i}g^-1__i .
let us now show that these constraints are sufficient in order to assign trivial holonomies to all cycles of @xmath101 which are mapped by @xmath368 to trivial cycles .
consider a non - fundamental cycle in @xmath101 which is mapped by @xmath368 to the trivial cycle ( i.e. to the root @xmath42 ) .
any arbitrary cycle in @xmath101 can be described by a word @xmath455 in the fundamental cycles , which are given by the union set @xmath456 .
we can therefore write that [ step1 ] ( w\{__i;__i } ) = r. we therefore have to show that the constraints imply that @xmath457 . since @xmath368 is a homeomorphism and
@xmath458 is a bijection , using @xmath459 , equation is equivalent to w\{(__i);(__i ) } = w\{__i;_(_^-1((__i ) ) ) } = r. using @xmath460 and again @xmath459 we can now rewrite this equation into [ trivialw ] w\{__i;_(w_i\{__i } ) } = w\{__i;w_i\{_(__i ) } } = w\{__i;w_i\{__i } } = r. the last relation is written in terms of the fundamental cycles of @xmath34 , which generate a free group .
the word @xmath461 on the right - hand side of can therefore be reduced to the identity by applying successively the concatenation of paths @xmath462 . in terms of holonomies ,
the constraints imply that @xmath463 . by using these relations in the word @xmath464 and by following the same steps of reductions ( using @xmath465 ) as for the cycles
, we can show that the word @xmath464 indeed reduces to the unit element of the group .
the notion of refined tree allows us to define a splitting of the algebra of observables on a finer triangulation , into a set of coarser observables which can be mapped to the observable algebra on a coarser triangulation , and into a set of finer observables .
this split will preserve the symplectic structure and can therefore be used to define the coarse - graining of spin network states ( see for instance @xcite ) .
this circumvents a problem noted in @xcite and coined `` curvature - induced torsion '' , namely the fact that if one coarse - grains an entire region into a ( spin network ) node , the fluxes associated to the links adjacent to this note do not necessarily need to satisfy the closure ( or gau ) constraint anymore .
this problem is avoided here by making the splitting of the algebra at the ( almost ) gauge - invariant level .
here we use a tree to determine the ( almost ) gauge - invariant algebra of observables .
therefore , every region which is coarse - grained into a node is entered by at least one tree .
the flux associated to this tree is however not part of the observable algebra , so one can thus not keep track of the closure constraint after coarse - graining .
a splitting of the algebra of observables underlies also the framework of @xcite aimed at enlarging the lqg state space .
however , @xcite leaves open the treatment of the gau constraints .
it was noted there that this requires indeed to work with parallel - transported fluxes .
here we do provide a full description of the corresponding algebra and a splitting of this algebra into finer and coarser observables .
let us recall the rooted holonomy - flux algebra @xmath171 , given by holonomies @xmath172 and fluxes @xmath173 . here
@xmath174 is the holonomy along the tree going from the root to the source node of the leaf @xmath51 .
the poisson brackets between these variables are given by ( for @xmath176 ) ^i_,^j_=_,^ij_k^k_,^k_,g_=_,g_^k-_^-1,^kg_,g_,g_=0 , where the @xmath177 denote the generators of @xmath178 .
we are going to show that , when considering a refined triangulation @xmath99 , the algebra of ( almost ) gauge - invariant observables can be split into two mutually commuting ( with respect to the symplectic structure ) algebras @xmath466 and @xmath467 , which correspond respectively to coarse and fine observables .
let us choose a refined tree in the graph dual to the refined triangulation and , following the notation introduced above , denote by @xmath468 the set of oriented leaves @xmath469 which satisfy the condition @xmath470 .
furthermore , let us denote by @xmath471 the set of oriented leaves in @xmath407 which satisfy @xmath472 .
the set @xmath473 therefore contains all the links dual to the fine @xmath31-dimensional simplices of @xmath46 which coarse - grain to the coarser @xmath31-dimensional simplex of @xmath33 dual to @xmath51 . with this notation
, we can now define on the phase space @xmath474 the refined extension is an extension of @xmath170 , which is different from @xmath475 . ]
@xmath476 of a flux observable @xmath477 as [ fluxrefined ] __i _ _^-1(_i)_+__^-1(_i^-1)_()^-1 .
this definition is a priori not the same as that of the integrated fluxes @xmath478 introduced in @xcite for co - paths @xmath281 of @xmath31-dimensional simplices in @xmath33 . for the fluxes @xmath478 ,
the parallel transport of the elementary fluxes ( associated to the @xmath31-dimensional simplices building up @xmath281 ) is defined through the notion of a shadow tree in the shadow graph of the co - path @xmath281 , and therefore stays as close as possible to @xmath281 . here
however , the parallel transport of the fluxes used to define is along the sub - tree which remains inside a set @xmath391 of @xmath14-dimensional simplices .
fortunately , the constraints impose the flatness of the parallel transport inside the coarse simplices .
one can thus replace the parallel transport for the integrated flux by a parallel transport along a surface tree without changing its action on refined states ( classically and on the constraint hypersurface ) . for the sake of obtaining a splitting into mutually commuting subalgebras
, we will stick with the definition .
the changes in the parallel transport described above might lead to commutators that vanish ( only ) on the constraint hypersurface .
we are now ready to split the holonomies and fluxes on @xmath474 into coarse and fine observables . for this
, remember that with our definition of refined trees there is a natural separation of the set of leaves of @xmath389 into two subsets @xmath393 and @xmath479 .
the decomposition of the holonomy - flux algebra on @xmath474 consists in assigning different observables for the leaves in these subsets .
explicitly , the coarse observables in @xmath466 are given by [ hfcoarse ] ( g__i,__i ) , _i _ , with the fluxes @xmath476 defined as in .
depending on the type of leaf @xmath438 , the fine observables in @xmath467 are given by [ hffine ]
@xmath480 this separation of the holonomies follows from the constraints and the result of lemma [ lemma : words ] . with this separation , we now have the following result , which we prove in appendix [ proof : separation ] : [ lemma : separation ] the coarse observables and the fine observables form two mutually - commuting subalgebras of the rooted holonomy - flux algebra @xmath171 , which we denote respectively by @xmath481 and @xmath482
. we can now introduce the refinement maps @xmath483 acting on the hilbert spaces @xmath95 .
we have seen that given a tree @xmath484 and a refined tree @xmath387 , there is a subdivision of the set @xmath407 of leaves of @xmath389 into the two disjoint sets @xmath393 and @xmath430 .
the leaves in the first set are labelled with small latin indices , while that of the second set carry capital latin indices . for every leave @xmath438
, we have a constraint [ constraints2 ] _
i w_i\{g__i}g^-1__i , as introduced in .
this constraint can be used to to define the refinement map in the following way : [ def : refinement ] for a state @xmath485 , we define the action of the refinement map as [ refined ] _
, ()\{g__i;g__i } \{g__i;g__i } = \{g__i}_i(w_i\{g__i}g^-1__i , ) .
this definition means that in order to obtain the refined wave function @xmath486 , we first have to replace the variables @xmath487 with the variables @xmath488 associated to the leaves @xmath489 in the set @xmath490 ( this set resulted from a choice of one representative leave in the finer triangulation @xmath46 per leave @xmath208 of the coarser triangulation @xmath33 ) , and then multiply the state with the ( discrete measure ) delta functions peaked on the constraints @xmath491 .
one can equivalently replace any of the constraints @xmath492 in the argument of the delta functions in with constraints @xmath493 . in this way
, we match the splitting of the observable algebra associated to @xmath46 into coarser and finer observables as described in section [ splitting algebra ] .
we are now going to address an important point , which is the gauge - independence of the refinement maps . in the definition , the state and its refinement
are expressed with respect to a particular gauge - fixing , i.e. a choice of tree , a choice of refined tree , and a choice @xmath393 for the set of coarse leave representatives .
we are going to prove that our construction is independent from this gauge information , i.e. that gauge transformations commute with the refinement operations .
let us consider a change @xmath145 of tree , a change @xmath494 of refined tree , and with it a change @xmath495 of the set of coarse leave representatives .
this leads to a different basis for the fundamental cycles .
if we denote the two different basis and their splitting into coarse and fine cycles by @xmath496 and @xmath497 , the change of basis can be written as _
_i = w__i\{__j;__j},__i = w__i\{__j;__j}. therefore as discussed in section [ change of tree ] , the associated gauge transformation of the refined state is given by [ refrans ] ( )\{g__j;g__j } = \{w__i\{g__j;g__j}}_i(w_i\{w__i\{g__j;g__j}}w^-1__i\{g__j;g__j } , ) .
our task is to show that this state is equal to the state which we would obtain by applying first a gauge transformation @xmath145 and then the refinement operation .
we are going to study the two terms of this expression separately .
let us first consider the delta function factors .
these give only a non - vanishing value equal to @xmath126 to holonomy configurations which satisfy the constraints .
these constraints can be expressed with respect to different choices of trees and sets @xmath393 .
let us denote the constraints in the cycle basis @xmath143 by _
j = _ j\{g__j}g^-1__j . as shown below
, these constraints are independent and sufficient to determine the constraint hypersurface on which every `` finer '' cycle carries a trivial holonomy .
this property does not depend on the choice of cycle basis being used , and therefore we can write that _ i(w_i\{w__i\{g__j;g__j}}w^-1__i\{g__j;g__j } , ) = _ j(_j\{g__j}g^-1__j , ) .
let us now focus on the first factor on the right - hand side of .
first , notice that due to the delta functions we can proceed to the following replacement : \{w__i\{g__j;g__j } } \{w__i\{g__j;_j\{g__j}}}. now , the argument of the state on the right - hand side also gives the transformation law for the fundamental cycles in @xmath34 , which can be obtained by applying @xmath368 to the fundamental cycles in @xmath34. this means that _ _
i = w__i\{__j;_j\{__j } } implies for the cycles @xmath433 and @xmath498 with respect to the trees @xmath49 and @xmath137 respectively the relation _ _
i = w__i\{__j;_j\{__j } } = w__i\{__j}. therefore , we can finally write that ( )\{g__j;g__j } = \{w__i\{g__j}}_j(_j\{g__j}g^-1__j , ) , and the right - hand side corresponds to the state which would have been obtained by applying first a gauge transformation @xmath145 in the coarse triangulation and then refining using a refined tree @xmath499 and a choice of set @xmath500 .
this shows that the refinement commutes with gauge transformations , and that the refinement operation is independent , the delta function factor in leads to the same set of configurations on which the refined state does not vanish .
second , a different choice for the set of coarse leave representatives will also replace some leaves @xmath426 by @xmath444 in the argument of @xmath147 in . however , any @xmath444 that replaces a @xmath426 needs to satisfy @xmath501 . due to the delta function factors we can therefore replace back @xmath502 with @xmath488 . ] from the particular choice of coarse leave representatives .
we now prove another very important property of the refinement maps , which is that they satisfy a transitivity condition .
this is needed in order for the inductive limit construction of the continuum hilbert space to be possible .
we have the following result : [ lemma : transitivity ] the embedding maps @xmath503 defined in satisfy _ ,_,
= _ , . in order to prove this result , it is sufficient to show that it holds for the basis vectors , and for a choice of tree in @xmath34 and of refined tree in @xmath101 .
the trees define an orthonormal basis for the two hilbert spaces @xmath95 and @xmath108 respectively . in terms of these bases ,
the refinement operation is given by [ eq : transitivityofrefinements ] _ , _\{_}=_\{_ } , where @xmath504 is determined as follows : [ labelref ] @xmath505 \alpha_{\ell'}'&=\alpha_\ell\q&&\text{if $ \mathbf{p}_{\delta',\delta}(\gamma_{\ell'})=\gamma_\ell$},\\[5pt ] \alpha_{\ell'}'&=\alpha^{-1}_\ell\q&&\text{if $ \mathbf{p}_{\delta',\delta}(\gamma_{\ell'})=\gamma_\ell^{-1}$.}\end{aligned}\ ] ] now , let us consider three triangulations @xmath506 , a tree @xmath49 , a refined tree @xmath389 for @xmath49 , and a refined tree @xmath423 for @xmath389 . by virtue of lemma [ lemma : ttransitivity ] , @xmath423 is then also a refined tree for @xmath49 .
therefore , we have that @xmath507 with @xmath508 \alpha_{\ell''}''&=\alpha_\ell\q&&\text{if $ \mathbf{p}_{\delta'',\delta}(\gamma_{\ell''})=\gamma_\ell$},\\[5pt ] \alpha_{\ell''}''&=\alpha^{-1}_\ell\q&&\text{if $ \mathbf{p}_{\delta'',\delta}(\gamma_{\ell''})=\gamma_\ell^{-1}$.}\end{aligned}\ ] ] a similar statement holds for @xmath509 . by employing the transitivity of the projection maps @xmath368
described in lemma [ lemma : ptransitivity ] , this shows the transitivity of the refinement maps @xmath510 . @xmath511
with the refinement maps at hand , we can finally define the inductive limit hilbert space @xmath367 .
such an inductive limit construction is based on a partially - ordered directed set , which is here provided by the set of triangulations @xmath512 .
the refinement maps ) satisfy @xmath513 the inductive limit is defined as the following disjoint union on which an equivalence relation is imposed : _
= , where @xmath514 and @xmath515 are equivalent , i.e. @xmath516 , if there exists a triangulation @xmath517 such that @xmath518 . in words , the two states are equivalent if they become eventually equal under refinement . the inductive limit carries a linear vector space structure which is inherited from the vector space structure on @xmath95 together with the linearity of the refinement maps .
the inductive limit can also inherit the inner product from the hilbert spaces @xmath95 .
for this , we need the refinement maps to be isometric , i.e. to satisfy the condition [ isometry ] _ , (_1)|_,(_2)_ = _1|_2_=_1|_2 _ , which is also known as cylindrical - consistency for the inner product .
this condition makes the inner product well - defined on the equivalence classes @xmath519 $ ] of states defined above . in order to compute the inner product between two states @xmath514 and @xmath515 ,
one chooses a common refining triangulation @xmath517 and takes the inner product between the states @xmath520 and @xmath521 in @xmath522 .
the condition ensures that the result does not depend on the choice of the common refining triangulation @xmath517 .
let us therefore show that the refinement maps are indeed isometric .
it is sufficient to show this for a set of basis states . in ,
we have considered the following refinement of the basis states : [ eq : transitivityofrefinements2 ] _
, _\{_}=_\{_ } , where @xmath504 is determined by [ labelref ] @xmath505 \alpha_{\ell'}'&=\alpha_\ell\q&&\text{if $ \mathbf{p}_{\delta',\delta}(\gamma_{\ell'})=\gamma_\ell$},\\[5pt ] \alpha_{\ell'}'&=\alpha^{-1}_\ell\q&&\text{if $ \mathbf{p}_{\delta',\delta}(\gamma_{\ell'})=\gamma_\ell^{-1}$.}\end{aligned}\ ] ] this refinement operation does indeed leave the inner product between basis states invariant in the sense that [ innerproductc ] _ \{_}|_\{_}_= _ ( _ , _ ) = _ , _\{_}|_,_\{_}_ , where we have made explicit the triangulation on which the inner product is computed .
this gives an inner product which is well - defined ( with respect to the equivalence relation ) on the inductive limit hilbert space @xmath367 .
in equation of appendix [ spinrepsu2 ] , we have defined a measure functional @xmath523 which reproduces the inner product .
this measure is also well - defined on the continuum hilbert space @xmath367 .
it can be expressed via its action on basis states as _ ( _ \ { _ } ) = _ ( _ , ) . finally , let us note that the construction of the fully gauge - invariant states via the raq procedure in section [ raqpro ] can also be extended to the continuum hilbert space
. these fully gauge - invariant states are then linear functionals on ( a dense subspace of ) the kinematical hilbert space @xmath367 .
the basic argument why this construction is cylindrical consistent is that the stabilizer group associated to a label @xmath135 for a basis state @xmath134 does not change if this state is refined .
this follows from the action of the refinement on the labels which is detailed in : an @xmath79-tuple @xmath135 gets extended to an @xmath524-tuple @xmath525 by @xmath406 including entries equal to the identity group element , @xmath396 copying an entry @xmath117 a number of times , and @xmath526 copying the inverse of an entry @xmath527 a number of times .
therefore , any @xmath161 which leaves the label @xmath155 invariant under the adjoint action does also leave @xmath504 invariant .
in addition to having a well - defined inner product on the inductive limit hilbert space @xmath367 , we would also like to have well - defined operators @xmath528 acting on @xmath367 ( here and in what follows we will denote operators without using hats on top of the symbols ) .
assume that we have at our disposal a continuum operator @xmath528 , i.e. a prescription for how @xmath528 acts on states @xmath529 for any triangulation @xmath33 ( possibly by mapping these states to finer hilbert spaces ) . the condition of cylindrical - consistency for such an operator @xmath528 is then [ cco ] _ ,
= _ , . this condition guarantees that applying the operator to states in the same equivalence class @xmath519\in\mathcal{h}_\infty$ ] gives the same result .
we will comment more in section [ contop ] on how to obtain such continuum operators . here
, we are rather going to focus on the following question : given an operator @xmath530 and a finer triangulation @xmath46 , we would like to define an extension @xmath531 which acts on the finer hilbert space @xmath108 and satisfies [ consistency o ] _ , (_)_, = _ , _. one could hope that starting with a given @xmath532 and extending it to all possible finer triangulations , one will end up with a description of a continuum operator @xmath528 .
however , the difficulty is that the extension maps which we are going to define are in general not transitive .
transitivity can nevertheless be obtained if one has a consistent way to project a continuum operator onto the different @xmath95 .
we will discuss this option in section [ contop ] .
for this discussion it will be helpful to first understand how an operator on a fixed triangulation can be extended to a finer triangulation .
this is going to be the topic of the present section .
an operator @xmath532 determines the properties of any extension @xmath533 on the subspace of states which are cylindrical consistent over @xmath33 ( i.e. states which can be obtained via refinements from @xmath95 ) .
for instance , the spectrum of @xmath532 coincides with the spectrum of @xmath533 restricted to this subspace . for a discussion of the notion of ( extended ) observables in the classical theory , and the relation to symplectic reductions ,
we refer the reader to @xcite .
let us recall how the refinement maps are defined . for a state @xmath534 , the refinement map acts as [ refa ] _
, ()\{g__i;g__i } \{g__i;g__i } = \{g__i}_i(w_i\{g__i}g^-1__i , ) . in words
, this means that we first replace the variables @xmath487 associated to the leaves @xmath425 with the variables @xmath488 associated to the leaves @xmath426 in the set @xmath490 , and then multipliy the state with the ( discrete measure ) delta functions peaked on the constraints @xmath535 .
we are now going to discuss separately the extension of the holonomies , of the fluxes , and of the parallel - transported fluxes .
let us start the discussion with a holonomy operator @xmath536 .
we define ( one version of ) an extension of this holonomy operator as [ exthol ] _ , (f\{g__i } ) f\ { g__i}. this definition means that we just replace the arguments @xmath487 in @xmath206 with @xmath488 ( with @xmath537 ) , and then see the resulting @xmath206 as a function on @xmath103 which is constant in all arguments @xmath502 ( with @xmath538 ) which parametrize the finer holonomies .
it is straightforward to see that this refinement map satisfies the extension property .
indeed , we have that [ refinedhol ] _ , (f)\{g__i;g__i } & = f\{g__i}\{g__i}_i(_i^1\{g_ } , ) + & = _ , (f)_,()\{g__i;g__i}. the prescription for extending holonomy operators is not the only possible one . for example
, the following changes do not affect the action of @xmath539 on refined states , and therefore do not affect the extension property : 1 . one can multiply @xmath539 with any other function on @xmath103 which evaluates to the identity on the constraint hypersurface .
any of the arguments @xmath488 in the extended holonomy operator @xmath540 can be replaced by @xmath488 multiplied with any word in the constraints @xmath535 and @xmath541 ( or their inverses ) , i.e. one can proceed to the replacement g__i g__iw\{_i;_i^-1}. this second modification amounts to deforming the path underlying the holonomies .
the original and deformed path must however be in the same homotopy class of paths on @xmath62 .
this freedom in changing paths is essential for the construction of continuum observables as discussed in section [ contop ] .
as explained in section [ sec : qfluxes ] , we obtain right translation operators @xmath542 as quantizations of the exponentiated ( flow of the ) fluxes @xmath170 , and left translations as quantizations of the exponentiated ( flow of the ) fluxes @xmath543 . in section
[ splitting algebra ] , we have defined one possible extension for the fluxes , when seen as classical phase space functions .
based on the choice of refined tree @xmath389 in @xmath46 , the extended flux is given by [ fluxrefined2 ] __i _ _^-1(_i)_+__^-1(_i^-1)_()^-1 .
accordingly , we can define an extension for the operators @xmath542 as _ , (r^ _ ) _
_^-1(_i)r^___^-1(_i^-1)l^_. in order to see that this definition satisfies the extension property , first note that @xmath544 leaves the ( kronecker ) delta functions in the constraints invariant , i.e. [ cflux1 ] _ ,
(r^_)(_i , ) = ( _ i , )
. this relation follows at the classical level from the fact that the fluxes @xmath476 poisson - commute with either @xmath535 or @xmath541 ( they commute with both versions weakly ) . to see this in the quantum theory ,
consider for instance the case described by equation .
there , we have a constraint @xmath545 associated to a leave @xmath444 which is in the link pre - image @xmath546 of some leave @xmath425 .
it is therefore not included in any other link pre - image @xmath547 with @xmath548 .
for this reason , @xmath549 will not affect @xmath545 for any @xmath548 .
for the case @xmath550 , we have that _ , (r^__i)f(g__ig^-1__i ) = ( r^__ir^__i)f(g__ig^-1__i ) = f(g__ig^-1__i ) , for any function @xmath551 .
therefore , the refined fluxes leave in particular @xmath552 ( and also @xmath553 since the argument is only changed by a conjugation ) invariant .
similarly , one can go through the remaining cases .
furthermore , it is easy to see that the action of @xmath554 on @xmath555 equals the action of @xmath556 on @xmath557 .
since the refinement maps multiply the states with delta functions in the constraints , and then replace the arguments @xmath557 in the coarse wave function with @xmath555 in the finer wave function , we can now conclude that the extension map for the fluxes satisfies the condition .
again , we can change a given extension of a translation operator without violating the extension property
. a particularly important class of changes is to include a parallel transport for the fluxes , in the form ^_,(r^ _ ) _ _^-1(_i)r^g[](g[])^-1__^-1(_i^-1)l^g[](g[])^-1_. here @xmath558 $ ] is an association to @xmath397 of an holonomy along a cycle ( starting and ending at the root ) which is homotopy - equivalent to the trivial cycle in @xmath62 .
this means that @xmath558=\openone$ ] if the ( coarse - graining ) constraints @xmath559 are satisfied . in order to avoid ordering issues
, one can demand that for a fixed coarse leave @xmath282 , none of the parallel transports @xmath558 $ ] associated to leaves @xmath397 in the set @xmath560 ( i.e. the set of leaves that appear in the refined translation ) include any of the leaves in @xmath561 . indeed ,
the parallel transport along a surface tree , as defined in @xcite , does not include any links which cross this surface .
one can therefore define an extension which matches the definition for an integrated flux ( where the finer fluxes are parallel - transported along a surface tree ) given in @xcite .
the restrictions on the set of leaves which can appear in the parallel transport can however be weakened .
for this , one can choose an ordering of the elementary translation operators by numbering the leaves in @xmath562 , so that @xmath563 , and then order the product of translation operators from @xmath564 ( appearing on the right ) to @xmath565 ( appearing on the left end ) . the condition on a parallel transport
@xmath566 $ ] is then that it should @xmath431 be homotopy - equivalent to the trivial cycle in @xmath62 , @xmath432 not include the link @xmath567 itself , and @xmath436 be such that @xmath566 $ ] is left invariant by the product of translation operators associated to the leaves @xmath568 .
let us now close this section by discussing the refinement of the parallel - transported translation operator @xmath569 , where @xmath191 is a holonomy associated to a ( rooted ) cycle @xmath48 which does not include the link @xmath208 itself .
the parallel transport @xmath191 is a word in the holonomies @xmath570 and their inverses for @xmath548 , i.e. @xmath571 .
we define @xmath572 as the same word in the holonomies @xmath573 , i.e. @xmath574 .
we can then define the extension of a parallel - transported translation operator as [ extensionpf ] _ , (r^g_g^-1___i ) _
_^-1(_i)r^g_(g_)^-1___^-1(_i^-1)l^g_(g_)^-1_. note that @xmath572 does not include any leaves in @xmath560 , and that therefore there are no ordering issues .
let us now discuss the interpretation of the parallel transports @xmath572 in the finer triangulation .
the holonomy @xmath191 describes a priori a rooted cycle in the coarse triangulation .
but it can also be interpreted as giving a parallel transport , alternative to the one provided by the tree , for the flux based at the source node @xmath575 of @xmath282 .
let us write @xmath191 as @xmath571 .
then , this alternative path is going from the root @xmath42 to the source vertex @xmath576 of @xmath208 , and is given by t_r_i(0)w_\{t_r_j(1)^-1l_jt_r_j(0)}. here @xmath577 is the path along the tree from the root @xmath42 to the source node of @xmath578 , and @xmath579 is the path along the tree from the root @xmath42 to the target node of @xmath578 . in the refined triangulation , @xmath572 provides in the same way alternative paths for the parallel transports of the fluxes @xmath580 . for a fixed @xmath581 ,
the path starts at the root @xmath439 and ends at @xmath582 .
it is given by t_r(0)w_\{t_r_j(1)^-1l_jt_r_j(0)}. likewise , if @xmath583 , the path ends at @xmath584 , and is given by t_r(1)w_\{t_r_j(1)^-1l_jt_r_j(0)}. again , one can also choose alternative extensions for the parallel - transported translation operators @xmath585 , in particular by changing the parallel transports @xmath572 .
for each @xmath580 , one can replace @xmath572 with a ( @xmath397-dependent ) transport along an homotopy - equivalent ( in @xmath62 ) path to @xmath48 .
as before , in order to avoid ordering issues , one can demand that these new paths should not include any @xmath580 .
this condition can be weakened in a similar way as discussed previously for the translation operators .
in the previous section , we have discussed extensions of operators on a given triangulation @xmath33 to a refined triangulation @xmath46 .
this allows to define a consistent family of operators given on a sequence @xmath586 of finer and finer triangulations . by taking the operator on a very fine triangulation @xmath587 in this sequence
, it is also possible to extend this consistent family to all triangulations coarser than @xmath587 . for this
, we just have to first refine states from the coarser triangulation to @xmath587 , and then apply the operator @xmath588 to such states .
this can be repeated with an even finer triangulation @xmath589 , which then gives a consistent definition on all triangulations which are coarser than @xmath590 ( agreeing with the previous definition on triangulations where it has already been defined ) .
however , there is no guarantee that starting with an operator @xmath532 and then extending this operator to @xmath46 and @xmath517 independently will give a definition which agrees on a triangulation that can arise as coarse - graining of @xmath46 and @xmath517 respectively .
the challenge is therefore to extend a given observable @xmath532 to all triangulations at once in a consistent way .
let us first discuss one possibility which does lead to a consistent family of observables , but however _ not _ to a continuum observable in the usual sense ( although it can be defined on the inductive limit hilbert space @xmath367 ) .
given @xmath591 on @xmath592 we define extensions to any finer @xmath108 by [ nullextension ] ( _ _ 0)_ ( _ _ 0)_i(_i , ) , where @xmath535 are the ( coarse - graining ) constraints describing states in @xmath108 which arise as refinements from @xmath592 . here
@xmath593 is any of the allowed extensions of @xmath591 to @xmath46 . it does not matter which extension one chooses , since the operator @xmath594 is only non - vanishing on states which arise as refinements of @xmath595 .
the actions of different extensions @xmath593 do agree on such states .
we can also define @xmath596 for any @xmath33 which is not a refinement of @xmath595 , by first refining a state in @xmath95 to a common refinement of @xmath595 and @xmath33 , and then applying the prescription .
the family @xmath594 of observables constructed in this way satisfies the cylindrical - consistency conditions .
it therefore defines an operator on the inductive limit hilbert space @xmath367 .
the observables do however vanish on any state which is not cylindrical over @xmath595 ( i.e. which is not a refinement of a state in @xmath592 ) .
such operators do therefore not correspond to quantizations of `` continuum '' flux and holonomy functions .
the classical version they correspond to are rather such functions multiplied with a ( infinite ) product of delta functions imposing that curvature is vanishing almost everywhere .
let us emphasize again that , nevertheless , the observables are well defined .
likewise , their spectrum has a well defined meaning .
this holds in particular with respect to the interpretation of the inductive hilbert space construction as corresponding classically to a symplectic reduction @xcite .
that is , the spectrum of @xmath532 coincides with the spectrum of @xmath597 restricted to the subspace of states which are cylindrical over @xmath33 .
the spectrum of @xmath594 is given by @xmath598 .
this leaves us with the question of whether one can construct quantum observables on @xmath367 which would actually correspond to quantizations of continuum versions of flux and holonomy observables .
let us sketch such a construction here , starting with an holonomy operator . to discuss the holonomy operator
, we will assume that the path @xmath48 underlying the holonomy is piecewise - geodesic in the auxiliary metric on @xmath13 .
the holonomy operator is unambiguously defined on a triangulation @xmath33 if the underlying path is a path in the inside of the manifold @xmath62 , i.e. if it avoids the defects of this manifold .
we call the set of triangulations @xmath33 on which this is the case @xmath599 .
this is itself a partially - ordered directed set , and the corresponding inductive limit hilbert space gives a subspace @xmath600 of the full inductive limit hilbert space @xmath367 .
the holonomy operator can unambiguously be defined on this subspace . in the case
in which the path @xmath48 does hit the boundary of @xmath62 one needs to choose a regularization of the operator ( or rather of the underlying path ) .
any such choice of regularization corresponds to an extension of an operator from @xmath601 to a bigger subspace .
such extensions of a ( bounded ) operator from @xmath600 to a ( bounded ) operator @xmath367 always exist , but they are in general not unique . for the translation operator ( or the parallel - transported translation operator ) , the main issue is again to unambiguously define the paths underlying the parallel transports of the translation operators associated to the finer links . in particular , for the notion of integrated flux over a surface @xmath266 ( in @xmath16 spatial dimensions )
, we have to specify a parallel transport for all points on this surface to , e.g. , a surface root .
the complicated case is here @xmath16 .
indeed , for @xmath15 there is a canonical construction for the parallel transport of the fluxes , which are integrated along a one - dimensional `` shadow graph '' as explained in @xcite . for the following discussion ,
we therefore restrict ourselves to @xmath16 , and furthermore assume that the auxiliary metric is flat , so that the notion of piecewise - geodesic reduces to that of piecewise - linear .
let us consider the integrated flux observables associated to a surface @xmath266 , which is itself piecewise - linear and more precisely comes from the gluing of ( flat ) triangular surfaces @xmath268 .
the continuum parallel transport for this surface is specified as follows .
first , this parallel transport is required to take place in the surface @xmath602 which is infinitesimally shifted below @xmath266 ( with respect to the orientation of @xmath266 ) .
second , we choose in each triangle @xmath268 of @xmath266 an inner reference point @xmath603 , referred - to as a @xmath268-root , to which all the inside points of @xmath268 are parallel - transported along geodesics ( which are here straight lines ) . finally , we choose a tree , referred - to as an @xmath266-tree , in the graph dual to the triangulation of @xmath266 .
the root of this @xmath266-tree should agree with that of the @xmath268-roots @xmath604 .
furthermore , the tree should connect the remaining @xmath268-roots @xmath603 to @xmath604 as follows : the nodes @xmath603 of two neighboring triangles are connected by a geodesic restricted to @xmath602 , i.e. a possibly kinked line in the @xmath16 flat space , and a straight line if considered from the induced ( flat ) metric of the two triangles .
this description defines the parallel transport for the continuum observable .
we now have to construct a projection of this parallel transport to a given triangulation @xmath33 .
this triangulation should be sufficiently fine in order to support the surface @xmath266 , i.e. the triangular surfaces @xmath268 are unions of two - dimensional simplices @xmath605 of @xmath33 .
consider the cut of the surface @xmath602 with the triangulation @xmath33 . by shifting @xmath266
infinitesimally to @xmath602 , we avoid the potential curvature singularities ( which in @xmath16 are along line defects ) along the edges of the triangulation @xmath33 ( remember that we also assume that the edges of the triangulation are geodesics , i.e. straight lines , and that the triangles of the triangulation are minimal surfaces , i.e. flat ) .
the edges of @xmath602 might now be `` thickened '' , since it could happen that more than two ( downside ) tetrahedra hinge on a given edge of @xmath266 .
we can however ignore this thickening since the path of the parallel transport across these additional tetrahedra is unambiguous .
what needs to be clarified is the path for the parallel transport across the two - dimensional triangulation of @xmath266 which is induced by @xmath33 .
this induced triangulation @xmath606 of @xmath266 will be a refinement of the original triangulation of @xmath266 into triangles @xmath268 , as shown on figure [ fig : subdivision ] .
we have to specify the parallel transport paths on @xmath607 , where @xmath608 denotes the vertices of @xmath606 .
we assume that the triangulation @xmath33 is such that none of the @xmath268-roots @xmath603 coincide with a vertex of the induced triangulation or lies on an edge of this induced triangulation .
we furthermore assume that the vertices of @xmath608 do not touch any of the paths along the @xmath266-tree .
therefore , we are just left with the task of specifying the tree for the parallel transport inside a given triangle @xmath268 .
we construct the parallel transport on @xmath609 as follows .
since the @xmath268-root @xmath603 lies in the inside of one of the induced triangles @xmath605 , the parallel transport for a point @xmath242 of this triangle @xmath610 to @xmath603 is therefore uniquely specified along the geodesic from @xmath242 to @xmath603 .
let us consider some other induced triangle @xmath611 of @xmath268 .
then we can have the following two cases : 1 .
the parallel transports along straight lines of all the points in @xmath611 are homotopy - equivalent in @xmath607 .
this means that for any two points @xmath612 and @xmath613 lying inside this induced triangle , the piecewise - geodesic path from @xmath603 to @xmath612 , then to @xmath613 , and back to @xmath603 , is homotopy - equivalent to the trivial cycle .
we can then choose any of the inside points @xmath242 , and adopt as a parallel transport for all points of this triangle a representative path which crosses through all the triangles which are crossed by the path from @xmath242 to @xmath603 .
there are points @xmath612 and @xmath613 of the inside of the induced triangle for which the piecewise geodesic path from @xmath603 to @xmath612 , then to @xmath613 , and back to @xmath603 , is not homotopy - equivalent ( in @xmath609 ) to the trivial cycle . in this case
, we have to subdivide the triangulation @xmath33 further , until this situation eventually does not appear ( now with respect to @xmath614 ) . to see that this procedure does indeed stabilize after a finite number of steps ( and does not produce further and further subdivisions ) , consider the example of figure [ fig : subdivision ] .
there , we construct the subdivision of @xmath268 into regions which are homotopy - equivalently transported to @xmath603 . to this end , we need to draw straight lines ( appearing as dashed lines in the middle panel of figure [ fig : subdivision ] ) from the @xmath268-root @xmath603 to all the vertices @xmath608 , and continue these lines until they hit the boundary of @xmath268 .
we can ignore the parts of the lines which lie in @xmath610 ( which are in dashed green in figure [ fig : subdivision ] ) .
the remaining parts of the lines ( in dashed red ) will subdivide the other triangles @xmath611 into homotopy - equivalent regions .
we then need to subdivide the triangulation further , until these regions are unions of triangles @xmath615 of the refined triangulation .
this introduces the finely dashed red line in the middle panel of figure [ fig : subdivision ] .
however , note that the lines emanating from @xmath603 do not cross themselves , and therefore only form new vertices with the edges which are already subdividing the triangle @xmath268 ( this is the vertex @xmath616 in the middle panel ) .
these new vertices can in principle refine the notion of homotopy - equivalent regions , which would then require an iteration of the procedure .
all the new vertices are however exactly on the boundary of the homotopy - equivalent regions which have just been defined .
therefore , an iteration of this procedure is not necessary .
we now have a triangulation of @xmath268 which is such that any two points in the inside of any given triangle are homotopy - equivalent as far as the transport to @xmath603 is concerned .
this allows to define a sub - tree , which specifies the parallel transport for the triangle @xmath268 itself ( see the right panel of figure [ fig : subdivision ] ) . as a last step
, one needs to find a refined triangulation @xmath46 of @xmath13 which induces a triangulation of @xmath268 that coincides with the one we have just constructed . and
its induced subdivision by a triangulation @xmath33 .
the vertices @xmath617 and @xmath618 lead to a further subdivision into homotopy - equivalent regions with respect to the transport to @xmath603 , which is shown in the middle panel .
this subdivision is completed to a triangulation of @xmath268 , and therefore the triangulation @xmath33 of @xmath13 needs to be refined to a triangulation @xmath46 such that the induced triangulation for @xmath268 coincides with the one in the middle panel .
the right panel shows the ( surface ) tree for the parallel transport in the triangulation @xmath46 . ] in summary , if the above assumptions are satisfied , we will eventually be able to define the parallel transport on a possibly refined triangulation @xmath46 , which is required in order to specify the translation operator ( on @xmath108 ) .
we can also apply this translation operator to states in @xmath95 since we just need to refine these states to @xmath46 . the discussion on alternative extensions to @xmath46 for a given translation operator defined on a triangulation @xmath33 in section [ extflux ] included in particular the change of parallel transport .
therefore , we can adjust the parallel transport of a given extension to @xmath46 so that it matches the one given by the projection of the continuum translation operator .
consider the ( integrated ) flux associated to a fixed choice of a triangulated surface @xmath266 , a choice of @xmath268-roots , and a choice of @xmath266-surface tree . to define the corresponding translation operator on the continuum hilbert space , we first consider a partially - ordered subset of triangulations such that the assumptions regarding the reconstruction of an unambiguous parallel transport are satisfied for each of the triangulations in this subset .
we can then argue in a similar manner as for the holonomy operators , and define an integrated translation operator on the corresponding subspace of @xmath367 .
extensions of this operator always exist , but are again not unique .
we therefore obtain continuum operators which are quantizing the holonomies and ( the exponentiated flow of the ) fluxes .
this allows to consider the ( weyl form of ) the holonomy - flux algebra , or more precisely its commutation relations . due to the cylindrical - consistency of the operators
, one can evaluate the commutators between two operators on a triangulation @xmath33 on which both operators can be unambiguously defined .
due to our quantization prescription , one then obtains the expected result for the commutators between holonomies and translation operators .
in @xcite , we have also evaluated the poisson algebra of the fluxes explicitly .
note that this can not be performed in the al framework , since there the poisson bracket of two fluxes might lead to a more singular object , e.g. a flux only smeared over a one - dimensional path in @xmath16 spatial dimensions .
in contrast , in @xcite one obtains again ( parallel - transported ) fluxes smeared over @xmath31-dimensional surfaces as a result of the commutator between two fluxes .
the reason is that the main contribution to the commutators arises from the parallel transport and the basic fluxes .
therefore , all the parallel transport holonomies for a certain ( @xmath31-dimensional ) part of the flux surface contribute to the commutator .
the result of the commutator is therefore a flux which is smeared over this part of the surface ( see @xcite for details ) .
the complete consideration of these commutators requires to go through many separate cases , and we therefore leave the consideration of the quantum commutators for future work .
we have constructed in this paper a new realization of quantum geometry .
our construction being based on the holonomy and flux variables of loop quantum gravity ( lqg ) , it shares some features with the already existing ashtekar
lewandowski ( al ) representation , but , most importantly , exhibits crucial differences .
first of all , both the al and the bf representations provide a concrete realization of the idea of seeing ( quantum ) gravity as a topological field theory with defects . however , while in the al representation the defects are the carriers of the geometrical excitations themselves , which leads to complications when trying to construct semi - classical smooth geometries , in the bf representation the situation is reversed .
indeed , as we have seen , the bf representation is based on a vacuum peaked on locally and globally flat connections , and excitations on top of this vacuum are carried by defects encoding non - trivial local curvature ( while the conjugated intrinsic geometry is maximally uncertain ) .
second , the construction of the bf representation relies on the triangulation of the spatial manifold itself , and does not consider the ( arbitrary ) dual graphs as fundamental structures , as it is the case in the al representation . because of this , the notion of refinement and embedding of dual graphs in the al representation is replaced here by the refinement of triangulations and the associated notion of embedding of coarser triangulations into finer ones . this notion of embedding is crucial for the construction of a continuum inductive limit hilbert space .
as recalled in the introduction , a fundamental result of lqg , which is based on the al representation , is the existence of a continuum inductive limit hilbert space which carries a unitary representation of the holonomy - flux algebra and of the action of spatial diffeomorphisms .
we have shown here that this result can be extended to the bf representation as well , and that it is possible to construct an inductive limit hilbert space in which the hilbert spaces associated to triangulations carrying a finite number of curvature excitations are embedded .
the existence of this continuum hilbert space relies on several key technical points .
the most important one is that we need to work with exponentiated ( i.e. compactified ) fluxes , which is in turn equivalent to considering a discrete topology on the group labelling the curvature excitations .
this is the reason why the bf representation evades the f lost uniqueness theorem @xcite .
indeed , this theorem states that the al representation is unique given a number of assumptions , and one of these is that the fluxes have to exist as well - defined operators .
the introduction of a discrete topology on the group is one of the key features which leads to different properties for the geometric observables in the bf representation as compared to the al representation . in particular , the spectrum of the area operator , discussed in section [ sec : spectrumar ] ,
is bounded in the bf representation . for the structure group @xmath131
, the spectrum can be either discrete or continuous , depending on a parameter @xmath8 appearing in the definition of the area operator .
we expect a similar result for the ( fully gauge - invariant ) @xmath2 area operator . furthermore , there are different possibilities to implement the presence of the barbero immirzi parameter into the area operator , and it would be very interesting to see whether this parameter will continue to appear in the black hole entropy state counting , for which the area spectrum plays a crucial role @xcite .
the fact that the spectra of the operators can be different in different representations emphasizes also the point raised in @xcite , namely that the spectra of physical observables on the physical hilbert space might be quite unrelated to the spectra of ( kinematical ) observables on kinematical hilbert spaces .
further key differences between the al and bf representations are the different requirements for the behavior of operators under refinement . in the case of the bf representation
, we can introduce for the geometric operators a measurement scale in the form of a subdivision of the region to be measured into a triangulation .
this triangulation is attached to the operator and not the the state .
therefore , the area of each triangle in such a triangulation is bounded .
this still allows to achieve large areas by sufficiently subdividing a region into pieces . it will be important to investigate how various results relying on the al representation will change if one works with the bf representation instead .
for instance , the results of the black hole entropy calculation @xcite might change if one uses the bf representation . in this direction ,
the work @xcite is interesting since it uses a kind of mixture between the bf and al representations in order to accommodate the isolated horizon constraints describing a certain family of black holes @xcite .
the advantages of considering this new bf representation are manifold , and there are lots of interesting and important directions to be explored , as explained in the following points : * the bf representation allows to have a nicer geometric interpretation since the states are not based anymore on degenerate spatial geometrical excitations .
one should explore this interpretation in particular in @xmath619 spacetime dimensions , and its relation to simplicial geometry .
since the classical phase space on a fixed triangulation agrees with that of loop quantum gravity on a fixed ( dual ) graph , one will find that the space of geometries , if interpreted in a simplicial manner , does not agree with regge geometries @xcite , but rather defines twisted geometries @xcite .
however , the basis states are in fact not semi - classical , but rather squeezed states .
in particular , the spatial geometrical operators have maximal uncertainties , whereas the curvature operators are maximally peaked .
an interesting question is therefore to what extent the connection determines the four - dimensional geometry .
note that the construction of such states with curvature line defects was also considered in @xcite . *
the bf representation forces us to work with the gauge - covariant fluxes , i.e. the fluxes with parallel transport .
this is intimately related to the behavior of the fluxes under refining and coarse - graining .
it furthermore explains the effect of `` curvature - induced torsion '' @xcite , which can be ( easily ) understood as the fact that a curved triangle or curved tetrahedron does not satisfy the so - called closure constraints .
this effect leads to a difficulty for coarse - graining , in the sense of mapping spin networks to coarser spin networks .
as noted in @xcite , the coarse spin networks do not necessarily need to satisfy the coupling rules .
however , this issue is largely connected to not providing a clean splitting of the observable algebra , as for instance used in @xcite for the case of the standard non - parallel - transported fluxes . here
, we provided such a splitting in section [ splitting algebra ] , which allows for a clear geometric interpretation of coarse - graining in terms of which observables are kept and which are averaged over .
this also answers the question raised in @xcite of how may quanta there are in a quantum spacetime : the phase space factor associated to the observables averaged over counts the number of quanta .
* in @xmath620 spacetime dimensions , the bf representation constructed here shares many properties with the so - called combinatorial quantization scheme @xcite , which is based on the quantization of the space of flat connections on a manifold with defects . in this combinatorial approach ,
the number of defects is fixed , so there is no need to invoke a continuum limit and therefore to use a bohr compactification like we do here .
nevertheless , it would be interesting to construct an explicit maps between these approaches .
reference @xcite already provides a map between the observable algebras .
* since the roles of the holonomies and the fluxes are inverted as compared to the al representation , one can now work with the fluxes themselves , and therefore describe the coarse - graining of geometrical quantities .
this would be a completion of the proposal @xcite to use a ( non - commutative ) flux representation for loop quantum gravity .
in fact , in appendix [ appendix : spin ] , we discuss the spin representation and a non - commutative product ( which is basically the matrix product ) based on this representation .
one needs this product in order to express the inner product of two states as a functional applied to the ( non - commutative ) product of these states .
+ in this respect , it is also interesting to explore the full algebra of the exponentiated flux operators .
in contrast , the full commutator algebra of the flux operators can not be described within the al representation since it would lead to fluxes smeared over singular surfaces in the case where the flux surfaces cut each other .
it has been shown in @xcite that in the case of the bf representation the classical poisson algebra of the fluxes is well - defined and does not lead to such an appearance of singular surfaces .
* in this work , we have left largely open the question of how to impose the spatial diffeomorphism and the hamiltonian constraints .
let us first comment on the spatial diffeomorphism constraints . here , we expect that spatial diffeomorphisms will act by moving the defects around , i.e. by changing the embedding of the triangulation network which supports the defects .
this is basically the same action as in the al representation , and one would therefore use , as in the al representation , a group averaging procedure in order to define the ( spatially ) diffeomorphism invariant hilbert space . + in @xcite , we have provided a generator ( on phase space ) for this action in the case of @xmath620 spacetime dimensions . as part of the task of understanding the geometries encoded in the @xmath619-dimensional case
, it would be important to investigate also this case .
the question there is whether and how the diffeomorphism symmetry coincides with the bf translation symmetry .
this is related to a proposal by zapata @xcite to construct a theory of topological gravity as a bf theory with certain broken symmetries .
* let us now turn to the hamiltonian constraints .
we do believe that hamiltonian constraint operators can in principle be constructed , since the main regularization mechanism pointed out in @xcite should also hold in our case .
however , we expect that the problems with the constraint algebra @xcite will persist .
this is ultimately related to the problem of preserving full diffeomorphism symmetry if lattices are introduced , even if this happens only on an auxiliary level @xcite .
an advantage of the bf representation is however the nicer geometric interpretation , which can facilitate the discussion of these issues .
the work @xcite is also aimed at understanding the dynamics of spin foam gravity as a continuum theory , starting from bf theory . + an alternative to directly imposing the hamiltonian constraints is to use a discrete time dynamics @xcite , and then to consider the continuum limit .
this would in fact fully follow the philosophy of approximating the dynamics by using defects in a regge - like manner @xcite . a framework for describing a simplicial canonical dynamics has been described in @xcite .
the question of how to reconstruct the continuum limit has been considered in e.g. @xcite . in this continuum limit , one can also hope to restore diffeomorphism symmetry as exemplified in @xcite . the above points concern possible future work directions related to the bf representation . in addition , we believe that the techniques developed in this paper will also be helpful to construct further realizations of quantum geometry .
any topological field theory whose defects support the holonomy - flux algebra ( or an alternative observable algebra ) can serve as a starting point for such a construction , as explained in @xcite .
a main question for future research is therefore whether there exists any such topological field theories in the case of four spacetime dimensions .
positive indications have been found in @xcite by studying the coarse - graining of ( simplified ) spin foam models , and work is in progress to find new topological field theories starting from spin foam models @xcite .
again , there are a number of further directions to explore , including the following ones : * a long standing problem has been that of constructing a hilbert space based on the structure group @xmath7 within the al representation .
the main difficulty in trying to do so is the non - compactness of the group .
even if one finds a way to appropriately compactify the group , there is the problem of how to construct an invariant measure . in the case of the bf representation ,
we need to compactify instead the dual of the group , which leads to a discrete topology on the group itself .
therefore , the advantage might be that there is an invariant discrete measure on the group which might allow to construct a bf - like representation for a non - compact gauge group .
* an important question is whether there is a way to avoid the requirement of compactifying the configuration space ( which , as we have discussed , can be understood from the necessity to define defects as stable physical entities ) , possibly along the lines of @xcite . in this vein
, it would be interesting to understand the general properties of inductive limit hilbert spaces and their limitations , and in particular to revisit the f lost uniqueness theorem @xcite . in particular
, the question is what kind of representations are allowed if one relaxes the assumption of the existence of ( non - exponentiated ) flux operators ?
* one can also consider cases in which the dual of the group is already compact .
apart from finite groups , this is the case of quantum groups at root of unity . these describe ( at least in three spacetime dimensions , since the results are only conjectured for the four - dimensional case ) quantum gravity with a positive cosmological constant ( and a euclidean signature ) .
there is by now a large body of work concerned with the issue of implementing the cosmological constant or quantum groups @xcite .
the introduction of a quantum group changes the almost everywhere flat connection to an almost everywhere homogeneously - curved connection .
a corresponding description in terms of simplicial geometry is given in @xcite .
+ when replacing @xmath2 with its quantum deformation at root of unity , the first question is to understand the structure of the defects .
in the case of @xmath620 dimensions , the defects have been understood to be given ( in the context of higher categories ) by the drinfeld center associated to the topological theory ( here the tuare
viro @xcite topological quantum field theory ) , as explained for instance in @xcite .
therefore , the construction of a tuarev viro - based representation can be achieved by combining the methods of @xcite and the methods developed in this work @xcite .
for the four - dimensional case , recent developments have been achieved in @xcite . + using a quantum group at root of unity will change the results for the spectrum of the area operator .
the compactification of the spectrum for @xmath2 means that the eigenvalues have to wind around in the interval allowed by the bound .
this can lead to a continuous spectrum in the case of lie groups . in the case of a quantum group at root of unity ,
the bound is on the ( admissible ) spins themselves , which means that the number of eigenvalues is itself actually bounded .
this would prevent the spectrum from being continuous . in summary ,
the existence of this new representation should not come as a surprise , and in fact indicates the presence of unsuspected rich phase structures in discrete approaches to quantum gravity . in ordinary quantum field theory ,
the existence of different vacua is tight to the description of condensation mechanisms , along with phase transitions and symmetry breakings .
different physical situations might be easier to describe within different representations based on different vacua .
recent results @xcite indicate that spin foam models can have non - trivial phase diagrams , with phases corresponding to particular topological field theories .
given such a topological field theory , one can describe excitations on top of a vacuum , and then dynamics for these excitations . in this way
, it can be expected that each phase will correspond to a different realization of quantum geometry .
this will not only open up a rich field of research , but also enrich our methods for constructing and understanding quantum geometries .
we would like to thank abhay ashtekar , klaus fredenhagen , and aldo riello for helpful discussions .
this research was supported by 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 research and innovation .
bb was funded by project 4966/1 - 1 of the german research foundation ( dfg ) .
mg is supported by the nsf grant phy-1205388 and the eberly research funds of the pennsylvania state university .
in this appendix , we briefly recall how the first fundamental group @xmath64 can be obtained in the case of two- and three - dimensional manifolds without defects .
first , if @xmath13 is a closed and orientable riemann surface of genus @xmath125 , its fundamental group @xmath64 is given by the free group on @xmath621 generators @xmath622 , divided by a normal subgroup generated by the single relator @xmath623\dots[a_g , b_g]$ ] , where the commutator is @xmath624=aba^{-1}b^{-1}$ ] .
this can be shown by using the seifert
van kampen theorem to write @xmath625 , where @xmath626 is the bouquet obtained from the wedge sum of @xmath621 circles , and @xmath314 is the normal subgroup generated by @xmath627 . since this relator corresponds to a null - homotopic loop , we can write the presentation [ 2dpresentation ] _ 1 ( ) = a_1,b_1, ,a_g , b_g|[a_1,b_1]
[a_g , b_g]=. notice that this first fundamental group , although always being finitely presented , is not free for @xmath628 . for a three - dimensional compact , connected , and orientable manifolds @xmath13
, the first fundamental group can also be finitely presented .
this is most easily seen by using the fact that any such manifold admits a cw complex . indeed ,
if we denote by @xmath629 the two - skeleton of the cw complex of @xmath13 , then we have that @xmath630 , which can in turn be computed using the seifert van kampen theorem and the fact that @xmath629 is the union of the one - skeleton @xmath631 with all the two - cells of the cw complex .
more precisely , @xmath632 is always given by a free group @xmath633 on @xmath634 generators , and the presentation of the first fundamental group of @xmath13 is given by _ 1 ( ) = a_1, ,a_n_1|r_1, ,r_n_2 , where the relator @xmath635 is the homotopy class in @xmath632 of the boundary of the @xmath202-th two - cell ( with @xmath636 ) , and therefore a word in the generators .
this can equivalently be described in terms of the so - called spine of the manifold @xmath13 , which is a two - complex consisting of a single vertex , one - cells and two - cells .
if the manifold @xmath13 is closed , its spine can be obtained by collapsing it after first removing an open three - ball . in this language , the generators are in one - to - one correspondence with the one - cells of the spine , the relators are in one - to - one correspondence with the two - cells , and the explicit expression for a relator @xmath635 describes how the @xmath202-th two - cell is attached to the one - skeleton .
the case of the defected manifold @xmath71 is simpler to deal with than the case of a manifold without defects .
this is due to the fact that the graph @xmath34 , which is the one - skeleton of the simplicial complex dual to the triangulation @xmath33 of @xmath13 , is a deformation - retract of @xmath71 . as a consequence ,
their first fundamental groups are isomorphic , i.e. @xmath72 , which is in turn immediate to obtain .
indeed , it is well - known that the first fundamental group of a graph is free and generated by @xmath637 generators corresponding to its fundamental cycles @xcite . by using a maximal spanning tree to characterize the fundamental cycles of the graph
, we can therefore say that the first fundamental group of @xmath62 is freely generated by generators associated to the leaves of the tree . for the sake of completeness ,
let us comment on the counting of the number of generators for the first fundamental group in the case of a punctured riemann surface of arbitrary genus .
any such surface with @xmath242 ( thickened ) punctures can be decomposed as a connected sum @xmath638 , where @xmath639 has genus @xmath640 and one puncture , while @xmath641 has genus one and @xmath642 punctures . in this decomposition , the surfaces @xmath639 and @xmath641 are glued along one common puncture ( so that the total number of punctures is indeed @xmath242 ) , and @xmath643 . by the seifer van kampen theorem , we have that the first fundamental group of @xmath71 is given by the amalgamated free product [ freeproduct ] _
1()_1(a)*__1(ab)_1(b ) = _ 1(a)*__1(^1)_1(b ) .
each factor in this expression can easily be computed .
first , for a surface of genus @xmath125 with a single puncture , the first fundamental group is isomorphic to that of a wedge sum of @xmath621 circles , so we can write that @xmath644 , which is the free group @xmath645 on @xmath646 generators .
second , for a surface of genus one with @xmath242 punctures , the first fundamental group is isomorphic to that of a wedge sum of @xmath642 circles .
this implies that @xmath647 , which is the free group on @xmath648 on @xmath649 generators .
finally , we have that @xmath650 , which is the free group on a single generator . gathering these results
, we get that the amalgamated free product defining @xmath70 is generated by the generators of @xmath651 and @xmath652 , with one generator removed due to @xmath653 .
therefore , this leads to a free group on @xmath654 generators .
this result is consistent with the usual presentation which generalizes to the case of punctured surfaces and takes the form [ punctured presentation ] _ 1 ( ) = a_1,b_1, ,a_g , b_g , g_1, ,g_p|[a_1,b_1]
b_g]=g_1 g_p
. this presentation is written in terms of @xmath621 generators corresponding to the non - contractible paths in @xmath71 , and the @xmath242 generators corresponding to the paths around the punctures .
now , this presentation can be reduced by a so - called nielsen transformation which , since there is only one relation in the presentation , indeed leads to the free group on @xmath655 generator .
now , since for triangulated surfaces the euler characteristic is given by @xmath656 , and since the punctures are `` piercing '' the faces ( which means that @xmath657 ) , we get that @xmath658 , which indeed corresponds to the number of leaves in a maximal spanning tree .
we explain in this appendix how to reconstruct the fluxes ( and operators ) associated to the branches from the ones associated to the leaves . to this end
we have to employ the gau constraints branches and @xmath659 leaves .
as far as the counting is concerned , it is therefore possible to reconstruct all the fluxes associated with the links @xmath82 from the knowledge of the fluxes associated to the leaves @xmath51 and @xmath57 gau constraints . ] . recall that the gau constraint for a node @xmath660 takes the form [ gauss1 ] _
l|l(0)=n_l+_l|l(1)=n_l^-1 0 , where the rooted fluxes are given by _
l = g^-1_rl(0)x_lg_rl(0),_l^-1=g^-1_rl^-1(0)x_l^-1g_rl^-1(0)=-g^-1_rl^-1(0)h_lx_lh_l^-1g_rl^-1(0 ) .
the last definition implies that for branches @xmath661 we simply have @xmath662 , while for leaves @xmath663 we have @xmath664 , where @xmath665 . given a choice of tree
, one can successively solve for all the fluxes by starting from the end nodes of the tree , where the gau constraints each involve only one flux associated to a branch . by solving for these fluxes associated to the end nodes of the tree
, we can go to the nodes which are only one step away from the end nodes along the tree . again , for every node there is only one unknown branch flux .
one can then iterate this procedure and solve all the gau constraints except for the one at the root .
the fluxes associated to branches are then given by linear combinations of the fluxes associated to the leaves and to the inverse leaves . to give the result of this procedure in a compact form ,
let us denote by @xmath666 the set of end nodes of the tree which lie after the node @xmath667 when following the orientation of the tree .
for each end node @xmath668 in @xmath666 there is a unique path @xmath669 going from the target node @xmath667 of the branch to the end node @xmath668 .
considering the set of all such paths for a given @xmath50 , we can then denote by @xmath670 the set of leaves which have their source node @xmath671 along the paths @xmath669 , and by @xmath672 the set of leaves which have their target node @xmath673 along the paths @xmath669 . as suggested by the notation
, a choice of branch @xmath50 automatically determines the sets @xmath670 and @xmath672 ( which can have a non - empty intersection ) . the flux is then given by _
b = _ _ ( b)_+__(b)_^-1 . in this formula , each leaf @xmath51 can appear at most once with the orientation @xmath51 and at most once with the orientation @xmath674 .
the right translations corresponding to a branch can therefore be defined as r_b^=__(b)r_^__(b)l_^.
in most of this article we have worked in the holonomy representation for the wave functions . in this appendix
we discuss a group fourier transform with respect to a discrete measure on the group , which defines a spin representation .
the discrete group @xmath131 can be understood as arising from a bohr compactification of its dual group @xmath132 .
consider the space of almost periodic functions @xmath675 on @xmath132 .
almost periodic functions are constructed as finite combinations of functions from the set @xmath676 .
since there exists an inner product , we also take the norm completion of this space .
this bohr compactification inner product is given by _
1|_2=_t_|k|t_2(k ) , and it turns the functions @xmath677 into an ( over - countable ) orthonormal basis .
this basis can be used to transform states to the `` holonomy representation '' , which is here in terms of an angle @xmath225 .
this transformation is given by [ trafo1 ] ( ) = _ |=_t_kte^ik(k ) , from which we recover the following kronecker delta form of the basis : _ ( ) = ( , ) , where @xmath127 if @xmath128 mod @xmath223 and is vanishing otherwise .
the inverse transformation to is given by ( k ) = _ ( ) e^-ik ( ) , with a discrete measure @xmath164 .
the translation operator r^ ( ) ( + ) acts on a basis vector as @xmath678 , and therefore in the @xmath44-representation as multiplication r^(k ) = e^ik(k ) . we can define a measure _ ( _ )
= ( , 0 ) for the basis functions @xmath677 , from which one can recover the inner product in the @xmath44-representation as _
1|_2=_(_k_2 ) .
this measure is the analogue ( or rather the dual ) of the haar measure ( which underlies the ashtekar
lewandowski measure ) , which in the context of @xmath131 would be defined as _
( _ k ) = ( k,0 ) for basis functions @xmath679 in the @xmath365-representation . in order to define a spin representation for @xmath2
, we proceed as in the case of @xmath131 .
consider a suitable space of functions on the set \{j , m , n|j/2 , m , n =- j ,- j+1,
,j-1,j}. by analogy with the case of @xmath131 , we define as `` suitable '' functions the functions which arise as finite linear combinations of functions from the set \{d^j_mn()|(2 ) } , where @xmath680 are the spin-@xmath11 representation matrices .
we write these functions into the matrix form @xmath681 , and equip the space of functions with the inner product [ c12 ] _
1|_2=__j=0^((_1^(j)_2(j ) ) , where @xmath682 . in order to define this inner product ,
we have introduced the regularization factor [ nfactor ] ( ) = _ j=0^d_j^2 = ( 4 + 3)(2 + 2)(2 + 1 ) , where @xmath683 . in this inner product ,
the following functions are orthonormal : _
( j)_mn = j , m , n|_=(d^j_mn())^ , and with @xmath684 constitute a ( over - countable ) basis .
we can now define the transformation to the group representation as [ c15 ] ( h ) = _ h|=__j=0^(d^j(h)(j ) ) , from which we therefore obtain _ ( h ) = _ _ j=0^d_j(d^j(h^-1 ) ) = ( , h ) , with @xmath685 iff @xmath686 and vanishing otherwise .
the inverse transformation to is given by ( j ) = _
( h)(d^j(h))^(h ) , with the discrete measure @xmath164 on the group .
the translation operator r^(h ) = ( h ) acts in the spin representation as ( r^)(j ) = ( d^j())(j ) . we can now also attempt to define a measure in the spin representation space which would recover the inner product
however , this can only be accomplished with a non - commutative product .
we therefore define a measure [ measurebf ] _ ( _ ) = ( , ) , and introduce a non - commutative ( matrix ) product for the basis states via ( _ _ ) ( j ) = _ ( j ) , which we extend by bi - linearity .
we can then express the inner product as _
1|_2=_(_1^_2 ) .
the spin representation discussed in the previous section can easily be extended from functions on @xmath1 to functions on @xmath78 .
for this , we consider states of the form \{j}_\{mn } = _ \{}c_\{}(_d^j_(_^-1)_m_n _ ) , where we have used the basis states _ \{j}_\{mn } = ( _ d^j_(_^-1)_m_n _ ) which are orthonormal in the inner product [ c25 ] _ 1|_2=__\{j}^(_1^\{j}_2\{j } ) . here
@xmath687 is a shorthand notation for ( _ 1^_2 ) = _ m_1,n_1,
,m_||,n_||()_m_1n_1 m_||n_||(_2)_m_1n_1 m_||n_|| .
we have therefore expressed the inner product as a summation over @xmath11 as well as on the magnetic indices @xmath688 .
these latter transform under the adjoint action , namely in a representation v_j_1v_j_1^* v _ j_||v_j_||^*. let us now choose a recoupling scheme and a unitary intertwining map u^\{j}_,j , l;\{m , n}:v_j_1v_j_1^* v _ j_||v_j_||^ * _ , jv_j , where @xmath689 is labelling the resulting representation , @xmath3 is the associated magnetic index , and @xmath690 summarizes the intertwiner ( i.e. intermediate spin ) labels . such an intertwining map can be built from @xmath691 symbols , which couple two representations @xmath692 and @xmath693 to @xmath694 . in this case ( and with certain phase conventions ) , the matrix elements of @xmath695 are real .
the condition of unitarity then means that u^\{j}_,j , l;\{m , n}u^\{j}_,j,l;\{m , n } = _ , _j , j_l , l , and [ c30 ] _ , j , lu^\{j}_,j , l;\{m , n}u^\{j}_,j , l;\{m,n } = _
\{m},\{m}_\{n},\{n}. we can then change our spin representation to ( \{j},,j)_l _ \{mn}u^\{j}_,j , l;\{m , n}\{j}_\{mn } , and also rewrite the inner product as a sum over the @xmath696 indices . to obtain the ( completely ) gauge - invariant part of a wave function @xmath697 , one would set @xmath698 ( and hence @xmath699 ) .
note however that a wave function with ( kronecker ) @xmath700 behavior would have vanishing norm in the ( rewritten ) inner product .
this is due to the scaling factor @xmath701 in ) .
however , it is not possible to change this scaling factor such that a sufficiently large class of wave functions have finite norm .
the reason for this is that the conjugacy classes @xmath702 require different scaling behaviors depending on which type of adjoint orbit they describe .
we can nevertheless apply the rigging map to a basis state in the recoupled spin representation : [ c32 ] ( _ \{})(\{j},,j)_l = _ g/(u)_(u)_\{m , n}(__m_n_)u^\{j}_,j , l;\{m , n}. note that this will not give a regular function in the indices @xmath703 since the @xmath698 is given by a ( discrete measure ) integral over a constant .
this integral only makes sense if one applies @xmath704 to a state in the kinematical hilbert space .
therefore , the physical inner product between two basis states can also be expressed as a sum over @xmath705 and @xmath706 .
the sum over @xmath705 leads , due to , to kronecker delta symbols in the indices @xmath707 , and the sum over these indices leads to the representation of the ( kronecker ) delta in the group elements , i.e. ( _ \{})_&=_g/(u)_(u)___j=0^d_j_j(u_u^-1_^-1 ) + & = _ g/(u)_(u)_(u_u^-1 , _ ) + & = ( [ \{}],[\ { } ] ) .
in this appendix , we prove that the translation operator @xmath219 acting on @xmath708 has a spectrum given by @xmath131 when the angle @xmath229 is irrational .
this actually holds when equipping @xmath131 with either a continuous or a discrete topology .
the difference between these two cases is just that the spectral measure for @xmath708 is continuous , while for @xmath709 it is discrete .
let us introduce the ket @xmath710 for @xmath711 .
note that this is _ not _ the usual eigenvector of the momentum operator for eigenvalue @xmath660 .
the translation operator @xmath219 acts as a right shift , i.e. @xmath712 .
let @xmath713 for @xmath251 .
we are now going to show that @xmath250 is in the spectrum of @xmath219 .
this is equivalent to saying that the operator @xmath714 does not have a bounded inverse .
first , if this operator has no inverse , we have the desired result and there is nothing to show .
if it does have an inverse , then we have to show that this inverse is not bounded . for this
, it is sufficient to show that there exists a sequence @xmath715 of vectors in @xmath249 such that @xmath716 but @xmath717 .
we now define this sequence .
it is given by @xmath718 , with c_n^(n ) = \ { cl e^in & , + 0&. .
this sequence satisfies @xmath719 , as well as ( r^-e^i)_n^2 = , where the right - hand side converges to zero .
this finishes the proof .
in fact , the ( formal ) limit @xmath720 does not converge in @xmath213 , but @xmath721 converges ( in the dual space of a suitable dense subspace ) to the generalized eigenvector corresponding to the point @xmath250 in the spectrum .
the whole spectrum is @xmath131 , and is continuous because the operator does not have any ( normalizable ) eigenvectors , but only non - normalizable ones .
in order to understand the subtleties underlying the notion of refined trees , let us focus on the example depicted on figure [ fig : refine - tree-1 ] .
there , we have a triangulation @xmath33 consisting of four triangles , a tree @xmath49 in its dual graph , and three leaves , @xmath722 , @xmath723 and @xmath724 , labeling the fundamental cycles @xmath725 , @xmath726 and @xmath727
. by subdividing the edge shared by the two upper triangles we get a finer triangulation @xmath46 , and we have represented on figure [ fig : refine - tree-2 ] three possible trees ( there are finitely many other possibilities ) in its dual graph @xmath101 . for each tree @xmath389 and the corresponding set @xmath407 of leaves , we can build paths @xmath454 describing the fundamental cycles by starting at the root , going along the tree until the source of @xmath397 , then going along @xmath397 , and back to the root along the tree .
then , we can project these paths with @xmath368 , and compare the resulting projections with the fundamental cycles described by @xmath49 .
let us do this for the three trees @xmath389 represented on figure [ fig : refine - tree-2 ] . 1 . with @xmath728 , we have that @xmath729 , @xmath730 , and @xmath731 .
this means that the projections of the fundamental cycles described by the leaves of @xmath728 correspond to the fundamental cycles described by the leaves @xmath53 of @xmath49 .
in addition to this , one can see that all possible paths in @xmath728 are mapped under @xmath368 to paths in @xmath49 .
2 . with @xmath732
, we have that @xmath729 , @xmath730 , and @xmath733 , while @xmath734 can be described in terms of the cycles @xmath726 and @xmath727 in @xmath34 .
however , at the difference with the tree @xmath728 , one can see that the tree @xmath732 has paths which under @xmath368 are not mapped to paths in @xmath49 .
3 . with @xmath735
, there are paths which under @xmath368 are not mapped to paths in @xmath49 , and the projections of the fundamental cycles do not correspond to the fundamental cycles determined by @xmath49 . indeed ,
although we have that @xmath729 , there is no leaf @xmath397 in @xmath735 such that @xmath736 or @xmath737 . with a tree
@xmath49 ( thick dashes ) in its dual graph , three leaves ( thin dashes ) labeling the three fundamental cycles , and a root @xmath42 ( thick node ) .
we have omitted the orientation of the simplices and of the dual graph for the sake of clarity . ] with three different trees in its dual graph . ] with this example , we see that the different possible trees which can be chosen in the refined graph @xmath101 do not characterize in an equivalent manner under @xmath368 the paths and fundamental cycles described by the coarser tree @xmath49 .
let @xmath47 , and @xmath49 be a maximal tree with a refined tree @xmath389 .
then @xmath389 is a result of the construction procedure described in section [ refinementtrees ] . in order to prove this result , we first look at a single @xmath14-dimensional simplex @xmath738 .
consider the edges of @xmath389 which start and end in this simplex , i.e. which start and end at vertices dual to @xmath14-dimensional simplices @xmath739 .
this collection gives a sub - tree of @xmath389 , which is connected due to property @xmath396 of a refined tree .
next , consider a @xmath31-dimensional simplex @xmath404 dual to a branch in @xmath49 .
because of @xmath406 , @xmath405 contains at least one @xmath31-dimensional simplex dual to a branch in @xmath389 .
if it did contain more than one , one could construct a closed loop in @xmath389 because of property @xmath406 .
this means that @xmath405 contains precisely one branch in @xmath389 .
all that remains to show is that the branches which we have singled out so far are the only ones in @xmath389 .
but the only edges which we have not considered yet are the ones which are mapped under @xmath368 to a leaf in @xmath49 .
so none of these edges can be a branch , because of property @xmath406 , and we are therefore done .
* lemma [ lemma : ttransitivity ] * ( transitivity of the refined trees)*. * _ let us consider three triangulations @xmath380 .
let @xmath389 be a refined tree for @xmath49 , and @xmath423 be a refined tree for @xmath389 .
then we have that @xmath423 is also a refined tree for @xmath49 . _
+ in order to prove this lemma , we need to show that the tree @xmath423 satisfies the properties @xmath406 and @xmath396 of definition [ def : refine - tree ] with respect to the tree @xmath49 .
property @xmath406 follows from result of lemma [ lemma : ptransitivity ] , and from the fact that the trees @xmath389 and @xmath423 are refinements of @xmath49 and @xmath389 respectively .
indeed , since @xmath740 and @xmath741 , by virtue of we also have that @xmath742 . for property @xmath396
, we have to show that any two nodes @xmath743 and @xmath744 , such that the dual simplices @xmath745 and @xmath746 are in @xmath747 for a fixed @xmath375 , can be connected by a path in @xmath423 without crossing the boundary @xmath748 .
let @xmath749 be the simplex in @xmath46 containing the simplex @xmath745 dual to @xmath750 , i.e. such that @xmath751 , and likewise for @xmath752 .
we have @xmath753 for @xmath754 .
then the following two situations can occur : 1 . if @xmath755 , the tree path from @xmath743 to @xmath744 is `` inside '' @xmath756 by virtue of property @xmath396 with respect to the pair @xmath757 of trees .
therefore , this path does also not cross the boundary of @xmath747 , which shows that @xmath396 holds in this case also for the pair @xmath758 .
2 . if @xmath759 , the path from @xmath743 to @xmath744 along @xmath423 projects under @xmath381 to a path along the tree @xmath389 , which is from the node dual to @xmath749 to the node dual to @xmath752 . by virtue of property @xmath396 for the pair @xmath760 of trees
, the path does not cross the boundary of @xmath761 .
together with the first point above , this shows that also the original path from @xmath743 to @xmath744 along @xmath423 does not cross the boundary of @xmath762 .
therefore , @xmath396 holds for the pair @xmath758 in general , which finishes the proof .
* lemma [ lemma : separation ] . * _ the coarse observables and the fine observables form two mutually - commuting subalgebras of the rooted holonomy - flux algebra @xmath171 , which we denote respectively by @xmath481 and @xmath482 . _
+ the proof of lemma [ lemma : separation ] can be obtained by a direct calculation .
first of all , it is clear that @xmath481 and @xmath482 each form a subalgebra which reproduces the poisson structure on @xmath763 .
the nontrivial statement is to show that these subalgebras are mutually commuting .
since holonomies are always poisson - commuting , the holonomies in @xmath481 have vanishing poisson brackets with the holonomies in @xmath482 .
in addition to this , since the leaves @xmath426 and @xmath444 are always distinct , the holonomies in @xmath481 commute with all the fluxes in @xmath482 .
let us now turn to the commutation relations between the fluxes in @xmath481 and the holonomies in @xmath482 . in , in addition to the leaf @xmath426 which always exists in @xmath764 , we have a leaf @xmath444 in the same pre - image .
therefore , for a given @xmath208 the poisson bracket between @xmath476 and @xmath535 contains two contributions , which are in turn vanishing since ( __i)^k,_i=^k__i+^k__i , g__ig__i^-1=g__i^kg__i^-1-g__i^kg__i^-1 = 0 .
similarly , for we have that ( __i)^k,_i=^k__i+^k_(_i)^-1,g__ig__i = g__i^kg__i - g__i^kg__i = 0 .
for , since the leaf @xmath444 does not belong to the pre - image of any leaf @xmath208 , it does not appear in the coarse observable @xmath476 and therefore the poisson brackets @xmath765 are identically vanishing .
this shows that the fluxes in @xmath481 commute with the holonomies in @xmath482 .
finally , let us look at the commutation relations between the fluxes in @xmath481 and the fluxes in @xmath482 . for , since @xmath766 , the unique contribution coming from this finer leaf to will be a flux @xmath767 .
however , because of the reversed orientation of the leaf in , the poisson brackets are vanishing since @xmath768 .
a similar reasoning applies to . for , since @xmath769 , the flux @xmath770 never appears in and the poisson brackets are therefore identically vanishing .
@xmath511 h. m. haggard , m. han , w. kamiski and a. riello , `` sl(2,c ) chern simons theory , a non - planar graph operator , and 4d loop quantum gravity with a cosmological constant : semi - classical geometry '' , ( 2014 ) , ` arxiv:1412.7546 [ hep - th ] ` . c. meusburger and b. schroers , `` the quantization of poisson structures arising in chern simons theory with gauge group @xmath771 '' , adv .
* 7 * , 10031043 ( 2004 ) , ` arxiv : hep - th/0310218 ` .
m. campiglia , c. di bartolo , r. gambini , j. pullin , uniform discretizations : a new approach for the quantization of totally constrained systems " , phys . rev .
* d74 * ( 2006 ) 124012 , ` arxiv : gr - qc/0610023 ` .
b. dittrich , `` the continuum limit of loop quantum gravity - a framework for solving the theory '' , in a. ashtekar and j. pullin , ed .
, to be published in the world scientific series `` 100 years of general relativity '' , ( 2014 ) , ` arxiv:1409.1450 [ gr - qc ] ` . b. bahr and b. dittrich , `` breaking and restoring of diffeomorphism symmetry in discrete gravity '' , proceedings of the xxv max born symposium `` the planck scale '' , wroclaw , 29 june - 3 july , 2009 , ` arxiv:0909.5688 [ gr - qc ] ` . | we construct in this article a new realization of quantum geometry , which is obtained by quantizing the recently - introduced flux formulation of loop quantum gravity . in this framework ,
the vacuum is peaked on flat connections , and states are built upon it by creating local curvature excitations .
the inner product induces a discrete topology on the gauge group , which turns out to be an essential ingredient for the construction of a continuum limit hilbert space .
this leads to a representation of the full holonomy - flux algebra of loop quantum gravity which is unitarily - inequivalent to the one based on the ashtekar
isham lewandowski vacuum .
it therefore provides a new notion of quantum geometry .
we discuss how the spectra of geometric operators , including holonomy and area operators , are affected by this new quantization .
in particular , we find that the area operator is bounded , and that there are two different ways in which the barbero immirzi parameter can be taken into account .
the methods introduced in this work open up new possibilities for investigating further realizations of quantum geometry based on different vacua . |
there is much speculation about the existence of tetraquarks .
most papers unfortunately do not distinguish between tetraquarks and simple meson - meson .
it is prudent , but less exciting , to examine conventional interpretations in terms of non - exotic @xmath15 , @xmath16 , @xmath12 or @xmath17 states with admixtures of meson - meson .
nuclei may be viewed as six quark combinations .
however , nucleon - nucleon phase shifts may be fitted in terms of meson exchanges .
the essential reason nuclear matter does not collapse is short - range repulsion due to the pauli principle . returning to meson spectroscopy
, the conventional @xmath18 states appear as octets and singlets ( or linear combinations ) .
jaffe has suggested that two coloured quarks can form a colour @xmath19 which is anti - symmetric , or @xmath20 which is symmetric @xcite .
then @xmath21 mesons can be formed as @xmath22 combinations of flavour and colour .
this neatly evades repulsion between @xmath23 and also accounts for the high masses of @xmath24 and @xmath25 . however , jaffe s scheme without meson - meson disagrees with observed branching ratios @xcite .
the ratio @xmath26 ( where @xmath27 are coupling constants ) requires that @xmath25 has a dominant @xmath28 component ; also the predicted ratio @xmath29 is much too small .
what may defeat jaffe s proposal is chiral symmetry breaking .
the pion and kaon are abnormally light , with the result that meson - meson configurations @xmath30 dominate over @xmath31 in @xmath32 , @xmath33 , @xmath24 and @xmath25 .
present measurements of branching ratios are not good enough to rule out some small admixture of @xmath34 ; further improvements in the branching ratios of @xmath32 and @xmath25 to @xmath35 , @xmath36 and @xmath37 above 1 gev would help greatly .
meson exchanges do make good predictions for @xmath37 @xcite and @xmath38 @xcite phase shifts up to masses where @xmath39 resonances appear .
they also predict correctly low energy @xmath40 and @xmath41 phase shifts .
there is no evidence for the @xmath42 representation predicted for @xmath43 combinations of flavour and colour . in a valuable review
, richard @xcite discusses the issues in terms of the flux - tubes within four - quark configurations .
there is a further feature relevant to meson - meson configurations .
the train of argument is as follows .
many mesons are observed at or very close to thresholds of opening channels .
well known examples are @xmath25 and @xmath24 at the @xmath35 threshold , @xmath44 ar the @xmath45 threshold and @xmath46 at the @xmath47 threshold within @xmath48 mev .
the mechanism for this synchronisation is rather fundamental @xcite .
the conventional form for the denominator of a resonance is @xmath49 where @xmath50 is lorentz invariant mass squared , @xmath51 , @xmath52 are coupling constants of open channels and @xmath53 are form factors .
however , the correct form for @xmath54 is @xmath55 where @xmath56 and @xmath57 here @xmath58 is the value of @xmath50 at the opening of the threshold and p denotes the principal value integral ; [ @xmath59 is the pole term from this expression . ]
the origin of eq .
( 2 ) is that amplitudes are analytic functions of @xmath50 , so that any change in the imaginary part of the amplitude must be accompanied by a change in the real part , or vice versa .
1 sketches the behaviour of @xmath60 and @xmath61 at an s - wave threshold , @xmath35 in this example .
the value of @xmath62 is large and peaks exactly at the @xmath63-wave threshold .
it acts as an attractor and can explain why @xmath25 and @xmath24 lie very close to the @xmath35 threshold .
the present limitation in calculating @xmath64 accurately is that form factors @xmath65 are poorly known .
incidentally , eq . ( 2 ) is equivalent to solving schr " odinger type equations , a procedure adopted by many authors .
these solutions are explicitly analytic .
it is also included in the model of van beveren and rupp , because their amplitudes are constructed algebraically to be analytic @xcite .
-12 mm -6 mm the hamiltonian for a @xmath18 state decaying to meson - meson obeys @xmath66 @xmath67 describes short - range @xmath18 components and @xmath68 refers to ingoing and outgoing mesonic channels and must include @xmath69- and @xmath70-channel meson exchanges ; @xmath71 accounts for the coupling between them due to @xmath50-channel decays . the eigenfunction @xmath72 becomes a linear combination of @xmath18 and meson - meson .
the latter is not an ` optional extra ' ; if a resonance decays to meson - meson , that component is a necessary part of the wave function , and indeed plays a vital role .
this is the basis of a large number of papers by oset and collaborators , enumerating the attractive meson - meson contributions to a large number of mesons , e.g. @xcite ; their calculations include empirical short - range terms which may well simulate @xmath18 contributions . the form of eq .
( 3 ) is strictly analogous to the formation of covalent bonds in chemistry @xcite . according to the variational principle , the eigenstate minimises the eigenvalue .
the @xmath18 component is of short range .
mixing with meson - meson components at longer range lowers momentum components in the wave function and hence the energy eigenvalue .
the @xmath73 , @xmath74 , @xmath75 and @xmath76 are very narrow and it is a good approximation to view them as pure @xmath77 states . some commentators then argue that other mesons can not be @xmath12 or @xmath17 because they have unexpected meson - meson components .
that is wrong .
if a resonance is close to an opening threshold with the same quantum numbers , it must contain virtual components of that channel .
taking @xmath78 as an example , there are attractive long range mesonic components in the configurations @xmath79 and @xmath80 ; examples are @xmath81 and @xmath82 , which appear prominently in observed decays of x , y , z in the mass range 3872 to 3945 mev .
the observed meson - meson decays are in fact a signature of mesonic components in the wave function .
weinstein and isgur modelled the @xmath25 and @xmath24 in terms of meson exchanges @xcite , as did the julich group of janssen et al .
@xcite .
these general remarks set the scene for further details .
section 2 discusses @xmath1 in this light .
it might be the first clear tetraquark .
however , it may also be understood at least qualitatively as a @xmath17 state coupled strongly to @xmath83 , @xmath84 and @xmath85 channels which open in the mass range 10559 to 10826 mev and can de - excite to @xmath86 , @xmath87 .
section 3 discusses states popularly known as @xmath88 .
the particle data group @xcite labels them as @xmath89 , @xmath90 and @xmath91 ; the last of these is somewhat confusing in view of the fact that its average mass is now @xmath92 mev .
here @xmath93 will be used as a shorthand for @xmath89 .
all these states lie close to the @xmath94 threshold .
this leads to my suggestion that all of these states are @xmath95 @xmath12 @xmath96-states mixed with meson - meson .
spin - parity determinations of these states are presently lacking .
this appears to be due to the fact that only decays have been analysed ; that leads to ambiguities .
it is essential to analyse the full process of production and decay , as in the analysis of dalitz plots .
this leads to orthogonal matrix elements between all @xmath97 .
the ideal process for analysis is @xmath98 ; here @xmath99 and @xmath100 both have spin 0 , making the formulae rather simple .
formulae for this process were published in 2005 @xcite , but experimental groups have not taken advantage of spin information from decays of @xmath101 and @xmath73 ; that is essential because it determines spin alignments . with the full formulae
, my simulation ( which assumes uniform geometric acceptance ) suggests that @xmath102 events are sufficient to determine @xmath97 if only a single amplitude is present or dominant .
if there are more amplitudes , it gets harder , but it would still be valuable to know the outcome .
section 4 reviews the partial wave analysis using tensor notation in its simplest form .
the formulae can be used in fully relativistic form . however , for the mass range 3872 to 3945 mev , relativistic corrections are small and barely significant with present statistics .
two appendices discuss the tensor notation , also the simplest choice of axes and questions of how to do lorentz transformations and rotations of axes , if needed .
section 6 summarises conclusions briefly .
-52 mm -52 mm
the belle collaboration reports branching fractions for @xmath103 , @xmath104 and @xmath105 a factor @xmath106 larger than for @xmath107 @xcite @xcite .
ali , hambrook , ahmed and aslam interpret these branching fractions in terms of a tetraquark @xmath2 @xcite .
more recently , ali , hambrock and mishima @xcite predict for this configuration an intensity ratio @xmath108/\sigma [ \upsilon ( 1s)k^0\bar k^0 ] = 1/4.\ ] ] for a @xmath109 or @xmath83 composition , this ratio should be 1 , allowing a clear experimental test .
there is a natural explanation of the sizable branching fraction for @xmath110 in terms of diagrams like that in fig .
2 . the intermediate states are off - shell , allowing @xmath111 production , even though @xmath112 and @xmath113 are below the kinematic threshold for pionic decays .
the final step allows de - excitation of @xmath114 to lower lying @xmath115 channels with larger phase space .
many similar diagrams exist with intermediate @xmath99 , @xmath112 , @xmath116 and @xmath113 . via mixing with meson - meson configurations ,
the wave function of @xmath1 becomes @xmath117 however , it is not possible to calculate these branching fractions accurately for lack of experimental information .
the decay width of @xmath107 to @xmath118 is very weak , probably because it decays almost entirely to @xmath119 .
transitions between @xmath99 and @xmath120 are forbidden by angular momentum conservation .
channels involving real @xmath112 and @xmath113 are closed , so only virtual intermediate states contribute . however , once intermediate @xmath83 , @xmath84 , @xmath121 and @xmath122 channels are open , there are 20 alternative diagrams
. there may be complicated interferences amongst them .
the conclusion is that the observed @xmath123 decays of @xmath1 are not dramatically out of line with expectation , bearing in mind the increase in phase space
. what would be decisive is an accurate determination of the branching ratio of @xmath1 to @xmath124 and @xmath125 .
a technical detail concerns the fit to the @xmath126 angular distribution .
the @xmath126 amplitude contains a broad pole at @xmath127 mev . in some sets of data
it appears as a peak , e.g. in @xmath128 @xcite . in others
it appears as in elastic scattering . in principle
, it therefore needs to be fitted by an amplitude @xmath129/d(s)$ ] , where @xmath130 is the adler zero , 0.0097 gev@xmath131 and @xmath132 are coupling constants .
an algebraic parametrisation of the denominator @xmath54 is given in ref .
it is not correct to fit the data with the e791 parametrisation , a breit - wigner amplitude of constant width .
that has a phase variation in complete disagreement with @xmath37 elastic scattering and violates watson s theorem @xcite .
furthermore , it contains an unphysical pole just below the @xmath37 threshold .
it is well known to theorists that the opening of strong thresholds can move resonances by large amounts . as one example
, barnes @xcite has studied charmonium loop diagrams and concluded that mass shifts are potentially @xmath133 mev ; he finds that they tend to be similar for all states in a multiplet with similar radial excitation number @xmath134 but different orbital angular momentum @xmath135 , e.g. the @xmath95 @xmath136 and @xmath14 states .
the @xmath93 is identified clearly in belle data for @xmath137 by a strong @xmath138-wave component in @xmath139 with mass @xmath140 mev , @xmath141 mev .
it is now adopted by the pdg as the @xmath89 radial excitation of @xmath142 @xcite .
there are two observations still labelled @xmath91 by the pdg .
one is by babar in @xmath98 with @xmath143 mev , @xmath144 mev @xcite . the second , with less events
@xmath145 , is by belle in @xmath146 with @xmath147 mev , @xmath148 mev @xcite .
the mass difference from @xmath149 is @xmath150 mev , so it is quite possible this is the @xmath89 decaying to @xmath151 .
another possibility is the @xmath152 .
there is considerable scatter in masses predicted for these states .
a representative collection of recent predictions is shown in table 1 .
mass differences between @xmath11 and @xmath13 radial excitations are large , but can be affected by thresholds and form factors .
if the @xmath95 @xmath11 predictions can be trusted , at least approximately , this state has been pulled down in mass by a large amount .
the strong signal for @xmath151 at 3915 mev is a hint of mixing between @xmath153 and the @xmath82 channel and attraction to this threshold and/or the @xmath81 threshold .
.recent predictions of masses of @xmath95 @xmath154 states and @xmath155 @xmath156 in mev . [ cols="^,^,^,^",options="header " , ] the decay of the resonance @xmath157
is described by combined spins @xmath50 of @xmath101 and @xmath73 ( or @xmath158 ) .
the @xmath93 is likely to decay dominantly with @xmath159 , though @xmath160 is allowed in principle .
however , the centrifugal barrier factor for @xmath161 is quite strong .
table 2 exhibits blatt - weisskopf centrifugal barrier factors for the amplitude using a generous radius of interaction 0.73 fm , including the convolution of both particles in the final state ; ( average masses are used for @xmath138 and @xmath162 ) . for a smaller radius , the barrier factors are even smaller .
the @xmath163 barrier makes that amplitude negligible .
the primary objective of the analysis should be to determine the dominant amplitudes , hence should start with @xmath164 and @xmath165 , both with @xmath159 .
further @xmath161 amplitudes may then be introduced one by one with penalty functions which constrain their magnitudes unless the data really demand them .
this may be done , for example , by adjusting the penalty function to contribute @xmath166 or log likelihood in the range 14 .
-35 mm -6 mm a simple choice of axes is shown in fig .
3 with the @xmath9-axis along the direction of the @xmath73 and @xmath167 along the normal to the plane of k , @xmath168 and @xmath73 .
the formula for @xmath169 is a 4-vector @xmath170 given in lorentz invariant form in section 4.2 of @xcite .
this formula is easily evaluated in any reference frame , such as that of fig .
it includes naturally the well - known enhancement near the edge of the dalitz plot for its decay to @xmath171 .
the @xmath73 decay is simplest in its rest frame .
the effect of the lorentz transformation to the rest frame of x , y , z is small because the @xmath73 is heavy and therefore non - relativistic in that frame .
the gymnastics required for the lorentz transformation of its polarisation vector @xmath172 are given in appendix 2 .
for @xmath173 , the kaon in the production process carries @xmath174 . in the resonance rest frame
, this angular momentum is described by the 3-momentum @xmath175 of the kaon . in the decay @xmath176 , the spin 1 of the @xmath177
is represented by the vector @xmath178 , where @xmath179 are momenta in any reference frame . in the non - relativistic limit
the full matrix element reduces to @xmath180 .
there is a simple trick which can be used to do the spin average over @xmath181 .
this is to write @xmath182 and @xmath183 ; then intensities are obtained by taking the modulus squared of @xmath184 . for @xmath46 , @xmath161 in the decay is eliminated by the tiny width of the state . the threefold dependence on the lepton axis , the spin of the @xmath177 and the kaon direction is then unique and highly distinctive . for @xmath164 , the matrix element is @xmath185 where @xmath186 and @xmath181 are 4-vectors . without the last term , which is very small ,
the matrix element is just the scalar product of 3-vectors @xmath186 and @xmath181 . with this simplification ,
the intensity is @xmath187 , where @xmath188 is the angle between @xmath186 and the lepton axis .
the simplicity of this formula may actually make the @xmath21 state harder to identify than other amplitudes . for @xmath189 , the kaon carries @xmath190 and is described by a tensor @xmath191 here @xmath192 is the usual kronecker delta function and @xmath193 . for decays with @xmath159 ,
the combined spin of @xmath101 and @xmath73 is @xmath194 .
the decay is then described by the tensor @xmath195 in the resonance rest frame ; @xmath196 implies the summation over @xmath197 to 4 and takes the same general form as @xmath198 above .
the full matrix element is just the contraction @xmath199 .
it is very distinctive because of the dependence on angles of all of @xmath100 , @xmath186 and @xmath181 .
a simulation of these @xmath97 , assuming full geometrical acceptance , suggests that 50 events would give a separation at @xmath200 confidence level between @xmath21 , @xmath201 and @xmath165 if only one of them contributes a significant amplitude .
babar and belle between them have many more events than this , with the exception of @xmath202 .
if more than one @xmath97 contributes , the separation is obviously poorer , and an overlap between @xmath21 and @xmath165 could be a difficult case .
formulae for @xmath161 in decays are given in ref .
@xcite .
a distinctive feature of @xmath46 is that its mass lies within 0.3 mev of the @xmath203 threshold .
there is a simple explanation .
it is attracted to this threshold by the dispersive effect due to the opening of the threshold .
lee et al .
@xcite model the @xmath46 including exchange of @xmath111 and @xmath32 between @xmath204 and @xmath162 .
they solve the schr " odinger equation , including the @xmath94 channel ; this is equivalent to evaluating the dispersion integral of eq . ( 2 ) , since their solution is explicitly analytic .
one should not be misled into thinking that the effect of @xmath94 is small because few events are seen in @xmath205 ; there are few events because of the narrow width of @xmath46
. however , the attraction to the threshold does not involve the resonance denominator in eq .
( 2 ) ; it arises from the integral over all phase space for @xmath94 .
a parallel example is the deuteron .
its binding energy is not governed by its tiny width ; it comes from virtual meson exchanges over the whole left - hand cut .
present data have failed to distinguish between @xmath173 and @xmath206 for @xmath46 .
the key point is to check whether partial wave analysis is consistent with @xmath173 or not . if it is , the fact that the mass coincides with the @xmath94 threshold is significant information .
a state with @xmath207 couples to @xmath94 with @xmath208 and there is no obvious reason why it should lie close to this threshold .
in fact , predictions for the mass of @xmath209 in table 1 lie far below @xmath46 ; this is because spin - splitting for @xmath138 states is smaller than for @xmath96-states and the mass of the @xmath210 ( mostly @xmath211 ) is well known . for @xmath212 , formulae are more messy and are given in ref .
there are are in principle three amplitudes , but one requires @xmath213 , which is irrelevant for the narrow @xmath46 .
the others have @xmath214 and combined spins either @xmath215 or 2 .
a full partial wave analysis of belle data for @xmath202 will in general be ambiguous with only @xmath216 events .
however , this may be enough events to test the possibility that @xmath217 .
the spin information is carried by the vector @xmath218,\ ] ] where @xmath179 are momenta ( in any frame ) . with @xmath219 ,
production requires @xmath220 and decays to @xmath81 with @xmath214 , so only a single amplitude is present ; it has the distinctive feature of going through zero at the centre of the angular distribution for decays .
there is a missing @xmath14 @xmath12 state . from predicted masses in table 1 and
the fact that the mass of the @xmath155 @xmath14 state is slightly about the @xmath221 state , it seems likely that the @xmath14 state will lie above @xmath46 .
however , it is not easy to find .
it has charge conjugation @xmath222 , while @xmath151 has @xmath223 , so this decay is forbidden .
it can appear in @xmath94 .
however , it appears in that channel in the combination @xmath224 ( because of the minus sign associated with @xmath225 ) , while @xmath221 decays with the opposite relative sign . without interferences with a state of established @xmath226 , @xmath14 and @xmath221
are not separated in any particular charge state .
it is necessary to be specific about what tetraquark configuration is meant in every discussion of them .
meson - meson combinations mixed with @xmath39 are to be expected via @xmath69- and @xmath70-channel exchanges and also from the @xmath50-channel decay . the increase in branching fractions for @xmath0 as the mass increases from @xmath107 to @xmath1
may be understood qualitatively via the opening of real decays to @xmath83 , @xmath84 and @xmath122 channels in the mass range 10559 to 10826 mev , followed by de - excitation to @xmath227 . observed branching fractions are not seriously out of line with expectation .
the composition @xmath228 may be tested by measuring the intensity ratio given by @xmath229/\sigma [ \upsilon ( 1s)k^0\bar k^0)$ ] , which ali , hambrock and mishima show to be 1/4 for exotic tetraquarks ; for a @xmath83 composition or @xmath17 , this ratio should be 1 .
spin - parity analyses of x , y , z states are needed and can be done straightforwardly in several cases using formulae developed in ref .
it is suggested here that they can be explained naturally as the expected @xmath95 radial excitations of @xmath221 , @xmath11 and perhaps @xmath13 .
it is essential to make use of the full forms of matrix elements for production and decay for partial wave analysis .
i am grateful to prof .
george rupp for valuable comments on the points discussed here .
chung gives an introduction to tensor formalism @xcite .
foundation work was done by zemach @xcite .
the convention used here is that the lorentz metric @xmath230 is such that four - vectors are written as @xmath231 , i.e. @xmath230 has only diagonal elements ( -1,-1,-1,1 ) .
the decay of a scalar particle , e.g. @xmath232 , is described by its mass @xmath233 ; here @xmath234 are 4-vectors of the decay pions .
a spin 1 particle needs to be described by a vector constructed to be orthogonal to the scalar expression .
this is done , for example in describing @xmath235 , using @xmath236 where @xmath237 substituting ( 11 ) into ( 10 ) , @xmath238(k_1 + k_2)_\mu}{m^2_\rho}.\ ] ] the orthogonality between vector and scalar is demonstrated from the dot product of @xmath239 with @xmath240 : @xmath241 in the rest frame of the decaying particle , @xmath242 has diagonal elements [ -1,-1,-1,1]-[0,0,0,1]= -[1,1,1,0 ] .
this gives the important result that @xmath243 is a 3-vector in the rest frame of a resonance .
then @xmath243 describes the angular momentum carried by the particle .
tensor expressions for higher spins are built up as described by zemach and chung so that they are orthogonal to lower spins .
they are symmetric and traceless , i.e. diagonal elements add to 0 .
we shall need only the tensor @xmath244 for spin 2 .
suppose the lepton pair from @xmath73 decay has components @xmath245 , @xmath246 , @xmath247 in axes in the @xmath73 rest frame parallel to @xmath88 of fig .
this vector may be expressed as @xmath248 the spin @xmath172 of the @xmath73 is orthogonal to this lepton axis . in axes
x , y , z aligned along the lepton axis , @xmath172 is given by a vector @xmath249 and eventually it is necessary to average over @xmath250 . in x , y , x , axes , @xmath251 is given by the converse relation to eq . ( 14 ) with the result @xmath252 @xmath253 under the lorentz transformation back to the centre of mass frame , this becomes @xmath254 where @xmath255 and @xmath256 are the usual expressions for the lorentz transformation .
this approach is simpler than that given in ref .
@xcite .
99 a. ali , c. hambrock , i. ahmed and m.j .
aslam , phys .
* b * 684 , 28 ( 2010 ) .
jaffe , phys .
d * 14 * 267 and 281 ( 1977 ) .
bugg , eur .
j c * 47 * 57 ( 2006 ) .
i. caprini , g. colangelo and h. leutwyler , phys .
lett . * 96 * 032001 ( 2006 ) .
d. descotes - genon and b. moussallam , eur .
j. c * 48 * 553 ( 2006 ) .
richard , arxiv : 1012.1022 .
bugg , j. phys .
g : nucl . part .
* 35 * 075005 ( 2008 ) .
e. van beveren and g. rupp , annals phys . * 324 * 1620 ( 2009 ) .
d. gamermann _ et al .
j. a * 33 * 119 ( 2007 ) .
bugg , j. phys .
g : nucl . part .
phys . * 37 * 055007 ( 2010 ) .
j. weinstein and n. isgur , phys . rev .
d * 41 * 2236 ( 1990 ) .
g. janssen , b.c .
pearce , k. holinde and j. speth , phys .
d * 52 * 2690 ( 1995 ) .
k. nakamura et al . , [ particle data group ] , j. phys .
g : nucl . part .
phys . * 37 * 075021 ( 2010 ) .
bugg , phys . rev .
d * 71 * 016006 ( 2005 ) .
et al . _ [ belle collaboration ] , phys .
. lett . * 100 * 11200 ( 2008 ) .
i. adachi _
[ belle collaboration ] , arxiv : 0808.2445 .
a. ali , c. hambrock and s. mishima , arxiv : 1011.4856 .
d * 80 * 074004 ( 2009 ) .
m. ablikim _
b * 598 * 149 ( 2004 ) .
bugg , j. phys .
g : nucl . part .
phys . * 34 * 151 ( 2007 ) .
watson , phys . rev . * 88 * 1163 ( 1952 ) .
t. barnes , arxiv : 1003.2644 . s. uehara _
et al . _ [ belle collaboration ] , phys .
lett . * 96 * 082003 ( 2006 ) . b. aubert _
[ babar collaboration ] , phys .
lett . * 101 * 082001 ( 2008 ) .
s. uehara _
et al . _ [ babar collaboration ] , phys .
. lett . * 104 * 092001 ( 2010 ) .
t. barnes , s. godfrey and e.s .
swanson , phys .
d * 72 * 054026 ( 2005 ) .
d. ebert , r.n .
faustov and v.o .
galkin , phys .
d * 67 * 014027 ( 2003 ) .
li , c. meng and k - t .
chao , phys .
d * 80 * 014012 ( 2009 ) .
[ belle collaboration ] , phys .
lett . * 94 * 182002 ( 2005 ) .
eichten , k. lane and c. quigg , phys .
d * 69 * 094019 ( 2004 ) .
swanson , phys .
rept . * 429 * 243 ( 2006 ) .
b. aubert _
[ babar collaboration ] , phys .
d * 68 * 092001 ( 2003 ) .
p. pakhlov _
et al . _ [ belle collaboration ] , phys .
lett . * 100 * 202001 ( 2008 ) .
bugg , j. phys .
g : nucl . part .
phys . * 36 * 075002 ( 2009 ) .
t. aushev et al . , [ belle collaboration ] , arxiv : 0810.0358 .
lee , a. faessler , t. gutsche and v.e .
lyubovitskij , phys .
d * 80 * 094005 ( 2009 ) .
chung , phys .
d * 48 * 1225 ( 1993 ) . c. zemach , phys
* 140 * b97 ( 1965 ) ; _ ibid _ * 140 * b109 ( 1965 ) . | data on @xmath0 show a large increase in branching fractions near @xmath1 .
a suggestion of ali et al .
@xcite is to interpret this as evidence for a tetraquark , @xmath2 .
however , it may also be interpreted in terms of @xmath3 , @xmath4 and @xmath5 above the open-@xmath6 threshold , followed by de - excitation processes such as @xmath7 . in the charm sector
, a hypothesis open to experimental test is that @xmath8 and @xmath9 peaks in the mass range 3872 to 3945 mev may all be due to regular @xmath10 and @xmath11 @xmath12 states ( and perhaps @xmath13 and @xmath14 ) mixed with meson - meson . *
meson spectroscopy without tetraquarks * |
the angular clustering of faint @xmath11-selected field galaxies has been studied extensively ( e.g. , efstathiou et al . 1991 ; roche et al . 1993 , 1996 ;
brainerd , smail & mould 1995 ; hudon & lilly 1996 ; lidman & peterson 1996 ; villumsen , freudling & da costa 1996 ; woods & fahlman 1997 ) , and a prime motivation of these studies has been to investigate the nature of the faint field population .
in particular , it is possible to infer the effective correlation length of the sample and the rate at which clustering evolves from a combination of the amplitude of the angular autocorrelation function , @xmath1 , and the redshift distribution of the faint galaxies , @xmath12 .
these observations can then be used to link properties of the faint field population with samples of local galaxies .
while the exact interpretation remains controversial , it is generally accepted that overall @xmath1 is fitted well by a power law of the form @xmath13 ( although see infante & pritchet ( 1995 ) for evidence of a flattening in the power - law coefficient at faint limits ) .
here we investigate the clustering of faint galaxies and focus on the behavior of @xmath1 at small angular separations .
we obtain a clear measurement of @xmath1 on scales of @xmath14 whereas previous investigations have been largely limited to scales of @xmath15 .
additionally , we use the clustering properties of the galaxies to estimate the number of pairs of galaxies that are physically close to each other in space ( separations of @xmath7 kpc ) .
the data consist of deep @xmath11-band imaging of 11 independent fields that were obtained in good conditions with the low resolution imaging spectrograph on the 10-m keck - i telescope .
each of the @xmath16 fields is centered on a high redshift quasar with high galactic latitude ; however , the presence of the quasar in the field is irrelevant to the present investigation ( i.e. , the presence of a small group of galaxies at the redshift of the quasar will not influence the results below ) .
the galaxy catalogs are complete to @xmath17 and the apparent magnitudes of the galaxies have been corrected for extinction . in order to reduce the stellar contamination in the object catalogs , only objects with @xmath18 are considered in the analysis below .
there is , of course , some residual stellar contamination of the galaxy catalogs at faint limits and we estimate that to be : @xmath1916% ( @xmath20 ) , @xmath1913% ( @xmath21 ) , @xmath1911% ( @xmath22 ) .
the integral constraints vary little from field to field due to the use of the same detector in all cases as well as the lack of very large , bright galaxies in the fields .
to compute the angular clustering of the faint galaxies we use the landy & szalay ( 1993 ) estimator : @xmath23 where @xmath24 , @xmath25 , and @xmath26 are the number of unique data - data , data - random , and random - random pairs within a given angular separation bin .
regions of the frame where faint galaxy detection was either lower than average or impossible ( e.g. , due to the presence of bright stars and galaxies ) were masked out when computing @xmath25 and @xmath26 .
raw correlation functions ( uncorrected for stellar contamination or the integral constraint ) were determined for each of the fields , from which a mean correlation function was computed .
the results for the mean raw correlation function are shown in figure 1 , where the error bars show the standard deviation in the mean . from top to bottom , the panels show the results for objects with @xmath20 , @xmath21 , and @xmath22 , respectively . also shown
are the formal best - fitting power laws of the form @xmath27 ( solid lines ) and the best - fitting power laws of the form @xmath13 ( dashed lines ) .
the power laws in the figure have been suppressed by the appropriate integral constraints and no correction for residual stellar contamination has been applied . the number of pairs of galaxies that we observe to be separated by @xmath28 is larger than the number predicted by the fiducial @xmath13 power law ( i.e. , the power law that is typically obtained from measurements that have been performed on scales of @xmath29 ) .
this is consistent with the results of carlberg et al .
( 1994 ) and infante et al .
( 1996 ) who both found @xmath1 to have a higher amplitude on small angular scales ( @xmath30 ) than a simple inward extrapolation of @xmath1 as measured at large angular scales .
as yet , however , it is unclear whether the steepening of @xmath1 is due to the existence of a population of `` companion '' galaxies ( which are not seen at the present epoch ) or luminosity enhancement ( e.g. , due to interactions ) of intrinsically faint galaxies that are in pairs . in the absence of significant luminosity enhancement
, we can estimate the number of pairs of galaxies that are physically close to each other simply by using the following probability : @xmath31 ( e.g. , burkey et al .
1994 ) , where @xmath32 is the number density of galaxies brighter than the faintest member in a pair of galaxies that is a candidate for close physical separation , @xmath33 is the observed angular separation between the galaxies , and @xmath34 is the smallest separation observed between all detected galaxies ( @xmath35 in our data ) . using eqn .
( 2 ) we compute the number of pairs of galaxies for which @xmath36 and @xmath37 in our data .
additionally , we use monte carlo simulations ( in which the magnitudes of the galaxies are shuffled at random ) to calculate the number of pairs of galaxies that would have @xmath36 and @xmath37 simply by chance .
the latter step allows the removal of random superpositions from the estimate of the `` true '' number of close pairs in the sample . below @xmath28
there are fewer pairs of galaxies with @xmath36 and @xmath37 in the actual data than are expected in a random distribution ( i.e. , based on the monte carlo simulations ) , indicating that we are undercounting the very closest pairs due to blending of the images . using our measured @xmath1 , however
, we can correct the faint pair counts on scales @xmath38 and estimate the fraction of galaxies in our sample that are in truly close physical pairs . based on a simple extrapolation of the cfrs redshift distribution
, we expect that the mean redshift of our galaxies is @xmath39 and , hence , if @xmath40 , physical pairs of galaxies that are separated by @xmath41 will be within @xmath42 kpc of each other .
the best - fitting power law form of @xmath1 ( corrected for stellar contamination and the integral constraint ) then yields an estimate of the pair fraction at this physical separation of @xmath6 for @xmath43 , which agrees with the results obtained by carlberg et al .
( 1994 ) for the fraction of galaxy pairs with separations @xmath44 kpc at @xmath8 .
this suggests , therefore , that little evolution in the merger rate of galaxies occurred between @xmath9 and @xmath10 .
a significant amount of work remains to be done on this project , including a clustering analysis of 7 additional independent fields and a more rigorous study of the number of pairs of faint galaxies located at close physical separation .
financial support under nsf contract ast-9616968 ( tgb ) and a boston university presidential graduate fellowship ( cjl ) are gratefully acknowledged .
the observations were obtained at the w. m. keck observatory , which is operated jointly by the california institute of technology and the university of california .
data analysis was performed exclusively on the origin2000 at boston university s scientific computing & visualisation facility .
brainerd , t.g . ,
smail , i. & mould , j.r . , 1995 , mnras , 275 , 781 burkey , j. m. , keel , w. c. , windhorst , r. a. , & franklin , b. e. , 1994 , apj , 429 , l13 carlberg , r. g. , pritchet , c. j. , & infante , l. 1994 , apj , 435 , 540 efstathiou , g. , bernstein , g. , katz , n. , tyson , j.a .
, & guhathakurta , p. , 1991 ,
apj , 380 , l47 hudon , j.d . &
lilly , s.j . , 1996 ,
apj , 469 , 519 infante , l. & pritchet , c.j . , 1995 ,
apj , 439 , 565 infante , l. , de mello , d. & menanteau , f. 1996 , apj , 469 , l85 landy , s.d . & szalay , a.s . , 1993 ,
apj , 412 , 64 lidman , c.e . & peterson , b.a . , 1996 , mnras , 279 , 1357 roche , n. , shanks , t. , metcalfe , n. , & fong , r. , 1993 , mnras , 263 , 360 roche , n. , shanks , t. , metcalfe , n. , & fong , r. , 1996 , mnras , 280 , 397 woods , d. & fahlman , g.g .
, 1997 , apj , 490 , 11 villumsen , j.v . ,
freudling , w. , & da costa , l.n . , 1996 , apj , 481 , 578 | we present a preliminary measurement of the angular clustering of faint ( @xmath0 ) field galaxies in which we concentrate on the behavior of @xmath1 on small angular scales ( @xmath2 ) .
the galaxies are strongly clustered and @xmath1 is well - characterized by a power law of the form @xmath3 .
the best - fitting value of the power law index , @xmath4 , is , however , steeper than the fiducial value of @xmath5 , indicating that there are more pairs of galaxies separated by @xmath2 in our sample than would be otherwise expected . using the best - fitting form of @xmath1 , we estimate that @xmath6 of the galaxies are in physically close pairs ( separations @xmath7 kpc ) .
this is a factor of order 2 larger than local galaxy samples but comparable to galaxy samples with @xmath8 .
the mean redshift of our galaxies is of order 0.95 , and , therefore , our result suggests that there was little or no evolution in the merger rate of galaxies between @xmath9 and @xmath10 . |
the x - ray transient xte j2123058 was discovered by the _ rossi x - ray timing explorer all - sky monitor ( rxte / asm ) on june 27 , 1998 ( @xcite ) .
x - ray bursts were detected in pointed rxte observations made on june 27 and june 29 ( @xcite ) , indicating that the system contains a neutron star .
based on the burst profiles and time intervals between bursts , we classify the bursts as type i ( @xcite ) .
if this system is a neutron star low mass x - ray binary ( lmxb ) , as the data suggest , it is unusual because of its high galactic latitude ( @xmath3 ) . _ in this paper , we report on optical and rxte observations of xte j2123058 .
our results include the optical identification of xte j2123058 ( @xcite ) , the measurement of a 5.9573 hr optical photometric modulation during its outburst ( @xcite ) , which proved to be the binary orbital period ( @xcite ) , and constraints on the binary inclination .
we place a lower limit on the distance from an estimate of the spectral type of the optical companion and quiescent optical observations and an upper limit from the flux of the x - ray bursts .
we study the x - ray variability on time scales from 0.008 to 4096 hz and detect high frequency quasi - periodic oscillations ( qpos ) , which were also reported by homan et al .
we derive model dependent estimates of the neutron star spin period and the magnetic field strength of the neutron star .
finally , we study the relationship between the optical and x - ray flux during the decline from outburst to quiescence .
the x - ray flux of xte j2123058 during its 50 day outburst is shown in figure 1 . the asm light curve was produced from data provided by the asm / rxte teams at mit and at the rxte sof and gof at nasa s gsfc .
the peak asm flux is about 85 mcrab ( 1.5 - 12 kev ) .
five pointed rxte observations were made during the outburst , and the proportional counter array ( pca ) flux measurements for each observation are shown in figure 1 .
table 1 provides information about the pointed observations .
we performed spectral and timing analysis of the pca data .
figure 1 shows the v - band light curve for the nine nights we observed xte j2123058 during the x - ray outburst .
table 2 lists these observations along with an observation made on september 20 after the x - ray outburst .
we used the 1.3 and 2.4 meter telescopes at the michigan - dartmouth - mit observatory ( mdm ) , the 0.9 meter telescope at kitt peak national observatory ( kpno ) and the 0.9 meter telescope at cerro tololo interamerican observatory ( ctio ) .
the exposure times were between 30 and 600 s. the data were reduced using standard iraf photometry reduction software .
figure 2 shows two [email protected]@xmath12.5 v - band images of the xte j2123058 field . the top image is an average of three 5 minute exposures taken on june 30 .
we identified the optical counterpart as a star in the 1@xmath12 pca error circle ( @xcite ) which had brightened to v = 16.9 from its quiescent level near the limiting magnitude , r = 21 , of a digitized uk schmidt sky survey plate ( @xcite ; @xcite ) .
the bottom image in figure 2 shows a v - band image of the xte j2123058 field taken on september 20 with the source probably in quiescence .
we measured the optical position relative to six neighboring stars in the usno a1.0 astrometric catalog of the palomar observatory sky survey ( @xcite ) .
the position of xte j2123058 is r.a .
= 21h 23 m 14s.54 , decl .
= 5@xmath13 47@xmath12 52.@xmath149 ( equinox 2000.0 ) with an uncertainty of 1@xmath14 ( 68% confidence ) .
during outburst , the xte j2123058 optical flux is highly modulated . to determine the period of the modulation
, we produced a lomb periodogram ( @xcite ) .
a highly significant peak occurs at a period of @xmath0 hr .
figure 3 shows the v - band light curve from july 1 to july 16 folded on this period . from july 1 to july 16 , the x - ray and phase averaged optical flux levels were relatively constant , and the mean peak - to - peak amplitude of the modulation was about 0.75 magnitudes . from observations of standard stars made on june 30 ( @xcite ) , the mean v - band magnitude was 17.27 during this time .
it is likely that the optical modulation occurs because one side of the optical companion is x - ray heated . in this case
, high levels of modulation are expected only if the binary inclination is relatively high .
further evidence for high binary inclination comes from the dip in the light curve near an orbital phase ( @xmath15 ) of 0.5 , which is probably caused by a partial eclipse of the optical companion by the accretion disk .
zurita et al .
( 1998 ) observed partial eclipses at @xmath16 and also at @xmath17 later in the outburst .
the photometric period we observe has been confirmed by spectroscopic measurements , which indicates that it is the orbital period of the system ( @xcite ) .
the light curve for each of six nights was fitted with a sine function and the time of minimum light was found to within @xmath18 d ( 68% confidence ) . for each night
, the cycle count can be unambiguously determined , and we find that 172 orbits occurred between the first and last minima .
the ephemeris was found by fixing the period to the known value and performing a @xmath19 minimization with the time of minimum light as the only free parameter .
the orbital ephemeris is hjd ( @xmath20 ) + ( @xmath21)e ( 90% confidence errors ) .
the rxte data were searched for modulation at the orbital period by folding the light curve for each observation on the orbital period .
we do not find evidence for modulation even though the rxte observations provide complete phase coverage .
the absence of modulation puts an upper limit on the binary inclination ( @xmath22 ) . using the empirical relationship for systems with main sequence optical companions from patterson ( 1984 ) , the companion for a 5.9573 hr binary
is expected to be a k8 star . in the following ,
we assume that the spectral type of the optical companion is between m0 and k5 .
we derive an upper limit on @xmath22 by assuming that the spectral type of the optical companion is m0 or earlier , and the mass of the neutron star is 1.4 m@xmath23 .
the no - eclipse condition for the neutron star is @xmath24
cos@xmath25 , where @xmath26 is the radius of the optical companion s roche lobe , which we derive from the relation given in eggleton et al .
( 1983 ) , and @xmath27 is the binary separation .
the no - eclipse condition gives @xmath28 .
a lower limit on @xmath22 can be derived if we interpret the dip in the optical light curve at @xmath16 as being due to a partial eclipse of the optical companion by the accretion disk . assuming that a thin accretion disk fills the neturon star s roche lobe , the partial eclipse condition is @xmath29 cos@xmath30 , where @xmath31 is the radius of the neutron star s roche lobe .
if the optical companion s spectral type is k5 or later , the inclination is greater than 55@xmath13 .
the times of the six x - ray bursts observed during the rxte observations are listed in table 1 , and the light curves for the three brightest bursts are shown in figure 4 .
the bursts shown in figures 4a and 4b were observed during observation 4 , and the burst shown in figure 4c was observed during observation 2 .
the properties of the observation 4 bursts are considerably different from the four dimmer bursts .
their rise times are 4 s compared to about 15 s for the dimmer bursts , and the observation 4 bursts are double - peaked , which may indicate that they are radius expansion bursts ( @xcite ) . since the peak burst luminosity is less than or equal to the eddington luminosity ( @xmath32 ) , a distance upper limit can be derived . using equation 4.10b of lewin et al .
( 1992 ) and assuming a neutron star mass of 1.4 m@xmath23 and a neutron star atmosphere with a cosmic abundance of hydrogen gives @xmath33 erg s@xmath10 .
it should be noted that @xmath32 would be larger for a more massive neutron star or a neturon star atmosphere with a lower hydrogen abundance .
we find the peak bolometric flux by producing 1 s energy spectra near the peak of the burst and fitting the spectra with a model consisting of the persistent emission plus a blackbody .
for the bursts shown in figures 4a and 4b , the peak bolometric fluxes are @xmath34 erg @xmath9 s@xmath10 and @xmath35 erg @xmath9 s@xmath10 , respectively , giving a distance upper limit of 15 kpc . due to the high galactic latitude of xte j2123058 ,
its distance is of considerable interest . here
, we derive the distance from estimates of the spectral type of the optical companion and the quiescent magnitude of the source .
observations made on august 26/27 gave r=@xmath36 ( @xcite ) , and we measured r=@xmath37 on september 20 .
there was no evidence for x - ray heating of the optical companion during either observation . since x - ray heating was not observed and the source flux remained constant between august 26/27 and september 20 , the source was probably in quiescence . assuming that the companion is a k8 star ( patterson 1984 ) and that the quiescent emission is dominated by emission from the optical companion gives v = 22.52 and an absolute v - magnitude of 8.5 , from which we derive a distance of 5.5 kpc . in deriving the quiescent v - band magnitude from the quiescent r - band magnitude we use a@xmath38 = 0.30 and a@xmath39 = 0.24 to account for extinction ( @xcite ) .
we note that the derived value for v is consistent with our september 20 v - band measurement .
a range of spectral types from m0 to k5 gives a distance range of 4.5 to 9.1 kpc .
we consider 4.5 kpc as a distance lower limit since this method underestimates the distance if light comes from the accretion disk in quiescence and conclude that the xte j2123058 distance is between 4.5 kpc and 15 kpc . from the distance lower limit and the galactic latitude , we find that the source must be at least 2.6 kpc from the galactic plane .
this is unusual since the scale height for neutron star lmxb is only 1 kpc ( @xcite ) .
four optical bursts occurred during our observations .
on june 30 , a burst of at least 0.15 v magnitudes occurred during a 60 s exposure .
we observed two bursts during the time from july 1 and july 16 , which are shown in figure 3 .
one of these bursts produced increases of 0.38 and 0.10 magnitudes for two consecutive 30 s exposures and the other produced a 0.21 magnitude increase during a 60 s exposure .
the largest optical burst was observed during the first exposure taken on august 6 . as shown in figure 9 ,
the 120 s burst exposure was 0.47 magnitudes brighter than the following exposures .
although we speculate that these bursts are reprocessed x - ray bursts , there were no simultaneous x - ray observations during any of the optical bursts . to see if the data are consistent with this hypothesis , we compared the energy released during x - ray bursts to that during the optical bursts . for a burst where the v magnitude increases from 16.9 to 16.5 and lasts 30 s ,
the burst fluence in the v - band is @xmath40 erg @xmath9 .
the x - ray fluence for an average xte j2123058 burst is @xmath41 erg @xmath9 giving an optical to x - ray burst fluence ratio of @xmath42 .
for the persistent emission , we find l@xmath43/l@xmath44 ratios between @xmath45 and @xmath46 .
the similarity of the persistent and the burst l@xmath43/l@xmath44 ratios indicates that the optical bursts are likely to be reprocessed x - ray bursts .
to search for qpos , we made 0.5 - 4096 hz power spectra from the rxte data .
we produced 2 s leahy normalized power spectra ( @xcite ) using the available data in the 2 - 20 kev energy band .
for each rxte orbit , the 2 s spectra were combined and we looked for qpos in the combined spectra . for observation 1 ,
it is not possible to carry out this analysis because there are no high time resolution data .
we only detect qpos during observation 3 , which confirms the result of homan et al .
a qpo at @xmath47 hz is detected during the first orbit of this observation and a qpo at @xmath48 hz is detected during the second orbit .
the errors given in this section are 68% confidence .
qpos are not detected for the other orbits .
fitting the power spectrum with a model consisting of a constant plus a lorentzian , we find that the 849 hz qpo has a width ( fwhm ) of @xmath49 hz and a fractional rms amplitude of @xmath50 % , and the 1141 hz qpo has a width of @xmath51 hz and a fractional rms amplitude of @xmath52 % .
f - tests indicate that both qpos are detected at greater than 99% confidence .
qpos are observed during x - ray bursts from several systems ( @xcite ) .
we searched for , but do not detect , qpos in the brightest two x - ray bursts . with the detection of khz qpos in xte j2123058 , this source joins a group of 18 other neutron star lmxbs with high frequency qpos ( @xcite ) . for several of these systems , two high frequency qpos are observed simultaneously . according to the beat frequency model
, the qpo with the higher frequency corresponds to the keplerian frequency at the inner edge of the accretion disk and the qpo with the lower frequency corresponds to the beat frequency between the keplerian frequency and the spin frequency of the neutron star . in this picture ,
the difference between the two qpo frequencies is the spin frequency of the neutron star ( @xcite ) .
this interpretation is supported by the fact that , in several sources , the difference between the two qpo frequencies is approximately constant even though the qpo frequencies change ( see e.g. @xcite ) .
in addition to the detection of the 849 hz qpo during the first rxte orbit of observation 3 , there is marginal evidence for a qpo at @xmath53 hz .
the feature is significant at the 86% confidence level .
the detection is significant at the 92% confidence level if we use only the last 2240 s of the first orbit data rather than the entire 3344 s. the detection can be further improved to 99% confidence if the 4.6 - 20 kev energy band is used rather than the 2 - 20 kev energy band .
this is consistent with the result of homan et al .
( 1999 ) that the qpo strength increases with photon energy .
figure 5 shows the 4.6 - 20 kev power spectrum for the last 2240 s of the first orbit fitted with a model consisting of a constant and two lorentzians .
both qpos are detected at greater than 99% confidence .
the qpo frequencies are @xmath4 hz and @xmath5 hz with fractional rms amplitudes of @xmath54 % and @xmath55 % , respectively . according to the beat frequency model , the observed frequency difference of @xmath56
hz implies a neutron star spin period of @xmath6 ms .
however , we note that the validity of the simple beat frequency model is in question because in some sources the difference between the two qpos is not constant ( @xcite ; @xcite ) . in fact ,
psaltis et al . (
1998 ) suggest that the qpo separation is not constant for any source with two khz qpos .
for the first four rxte observations , we produced 2.520 kev pca energy spectra using the processing methods described in tomsick et al .
( 1998c ) . for observation 5 ,
the data were processed in the same way except that the background spectrum was estimated using a model for faint sources ( @xcite ) . only the photons from the top anode layers were used . as described in tomsick et al .
( 1998c ) , crab spectra were made to test the response matrices .
of the five proportional counter units ( pcus ) , the response matrix is best for pcus 1 and 4 .
since pcu 4 was off during observation 1 and part of observation 2 , the spectral fits were carried out using only pcu 1 in order to avoid instrumental differences between observations .
also , fluxes or spectral component normalizations were reduced by a factor of 1.18 so that the pca flux scale is in agreement with previous instruments ( @xcite ) .
energy spectra for the persistent emission were produced using data with 16 s time resolution .
time bins which overlap with x - ray bursts as specified in table 1 were excluded .
when the first two observations were made the source had not been optically identified and the source was about 0.2 degrees from the center of the field of view .
the june 5 , 1996 collimator response was used to correct for this pointing offset .
the spectra were fitted with a power - law model , a comptonization model ( @xcite ) and a model consisting of a blackbody plus a power - law . for each model
, the fit was calculated with the column density free and also with the column density fixed to the galactic value ( @xmath57 @xmath9 ) .
a power - law alone provides a very poor fit to the first four spectra , but for the fifth spectrum , a power - law with a photon index ( @xmath58 ) of 1.98 provides an acceptable fit ( @xmath59 = 36/45 with the column density free and @xmath59 = 37/46 with the column density fixed ) .
the fits to the first four spectra are improved by using a comptonization model or a model consisting of a blackbody and a power - law .
the fit parameters for these two models are given in table 3 .
for the first four spectra , comptonization models with temperatures ( @xmath60 ) between 2.5 and 3.0 kev and optical depths ( @xmath61 ) between 11 and 14 provide reasonably good fits to the data . for observation 5 ,
the comptonization model does not provide a better fit than the power - law model , and the comptonization parameters are very poorly constrained .
the column density is consistent with the galactic value for observation 5 if a power - law or a comptonization model is used . for observations 2 , 3 and 4 ,
f - tests indicate that the spectra require a column density different from the galactic value at greater than 90% confidence if a comptonization model is used .
the blackbody plus power - law fits give similar results for the first four spectra , but the fitted parameters are significantly different for observation 5 . for the first four spectra with the column density free , the blackbody temperatures are between 1.86 and 1.94 kev , the values for the photon index , @xmath58 , are between 2.5 and 3.1 , and the column densities are between 1.3 and 2.6 @xmath62 @xmath9 . for observation 5 ,
the blackbody temperature is 0.65 kev , @xmath63 , and the column density is consistent with the galactic column density . for observations 2 , 3 and 4 ,
f - tests indicate that the spectra require a column density different from the galactic value at greater than 99% confidence . near the end of the x - ray outburst from aql x1 in early 1997 , zhang et al .
( 1998 ) observed a sudden change in the hardness of the x - ray spectrum .
they interpret the change as the onset of the `` propeller '' mechanism ( @xcite ) . in this picture , as the mass accretion rate onto the neutron star drops , the size of the neutron star magnetosphere increases until it produces a centrifugal barrier for the accreting material .
zhang et al . ( 1998 ) fit the aql x1 energy spectra with a model consisting of a blackbody , a power - law and an iron line and find two main changes after the transition : the blackbody temperature decreases and the power - law index becomes harder . for xte j2123058 , the power - law plus blackbody fit parameters for observations 4 and 5 show a similar change .
the energy spectra for observations 4 and 5 fitted with a blackbody plus power - law model with the column density free are shown in figure 6 . between observations 4 and 5
the blackbody temperature changed from @xmath64 kev to @xmath65 kev ( 68% confidence errors ) and the power - law index changed from @xmath66 to @xmath67 . assuming that the change is due to the onset of the propeller mechanism , we can estimate the strength of the neutron star magnetic field using equation 2 of zhang et al .
( 1998 ) . for this estimate
, we assume 1.4 m@xmath23 and @xmath68 cm for the mass and radius of the neutron star . assuming that the neutron star spin period is 3.92 ms and that the critical x - ray luminosity for the transition is between the observation 4 and 5 levels
, the neutron star magnetic field strength is between 2 and 8 @xmath69 g , where @xmath70 is the source distance in units of 10 kpc .
for observations 2 to 5 , we produced 0.008 - 128 hz power spectra to study the timing properties of the system .
an rms normalized power spectrum was made for each 128 s interval using non - burst data in the 2 - 20 kev energy band . to convert from the leahy normalization to rms , we determined the poisson noise level using the method described in zhang et al .
( 1995 ) with a deadtime of 10 microseconds per event .
the background count rates were estimated as described in the previous section . for each of the four observations ,
the 128 s power spectra were combined and then fitted with a power - law model . for observations 2 and 5 , a power - law provides an acceptable fit to the data ; however , for observations 3 and 4 , we detect excess power above 5 hz . for observations 3 and 4 ,
the power spectra were fitted with a model consisting of a power - law plus a cutoff power - law .
this model has the form @xmath71 where @xmath72 , @xmath73 , @xmath74 , @xmath75 and @xmath76 are free parameters .
for all four observations , the fitted parameters are given in table 4 .
instead of reporting the values for @xmath72 and @xmath74 , fractional rms amplitudes are given for the two components . during the outburst
there were two significant changes in the timing properties of the source .
the first change occurred between observations 2 and 3 when excess noise above 5 hz appeared in the spectrum . for observations 3 and 4 , the fractional rms amplitudes ( 0.01 - 100 hz ) for the cutoff power - law component are @xmath77 % and @xmath78 % , respectively ( 68% confidence errors ) . for observation 2 ,
the 90% confidence upper limit on the fractional rms amplitude of a similar component is 1% .
the second change occurred between observations 4 and 5 . for observations 2 , 3 and 4
the fractional rms amplitudes are between 2.2% and 2.8% , while for observation 5 , the fractional rms amplitude is @xmath79 % .
the power spectra for observation 2 , observations 3 and 4 combined and observation 5 are shown in figures 7a , 7b and 7c , respectively . although a power - law provides an acceptable fit for observation 5 ( @xmath59 = 10.7/12 ) , the observation 5 power spectrum was re - fitted with the function given in equation 1 .
the fitted cutoff power - law parameters are @xmath80 and @xmath81 hz . the 90% confidence upper limit on the fractional rms amplitude of the cutoff power - law component is 21% .
we made a color - color diagram ( cd ) to see how the changes in the xte j2123058 timing properties are related to its spectral properties .
figure 8 shows the cd , where the ratio of the 3.5 - 6.4 kev to 2.1 - 3.5 kev count rates is plotted on the horizontal axis ( soft color ) , and the ratio of the 9.8 - 20 kev to 6.4 - 9.8 kev count rates is plotted on the vertical axis ( hard color ) .
the source occupied different regions of the cd during the outburst .
the cd suggests that xte j2123058 should be classified as an atoll source ( @xcite ) , and that the source was in the upper banana state ( ub ) during observations 1 and 2 , the lower banana state ( lb ) during observations 3 and 4 and the island state during observation 5 ( @xcite ) .
this interpretation is supported by the timing properties of the source during observations 2 to 5 . a change from the ub state to the lb state
is typically accompanied by an increase in the high frequency noise component ( which we characterize using a cutoff power - law ) .
also , the increase in the rms amplitude we observe between observations 4 and 5 defines the transition from the banana state to the island state .
that xte j2123058 is an atoll source is not surprising since x - ray bursts are commonly seen in atoll sources .
the timing analysis presented here causes us to question whether the change observed during the outburst decay is related to the onset of the propeller mechanism . based on its cd and the timing properties during the decay
, the behavior of xte j2123058 appears to be typical for atoll sources as the mass accretion rate onto the neutron star decreases .
it is not clear if the spectral and timing changes are due solely to the dynamics of the disk or due to the interaction between the disk and the magnetosphere .
the latter possibility may suggest that the onset of the propeller mechanism causes the change from the banana to the island state in atoll sources .
a argument against this is that the x - ray timing properties for black hole systems , where the compact object does not have an intrinsic magnetosphere , are similar to atoll source timing properties ( @xcite ) .
we observed xte j2123058 in the optical at x - ray flux levels from quiescence to @xmath82 erg @xmath9 s@xmath10 ( 1.5 - 12 kev ) .
our optical observations provide sufficient phase coverage to study the shape of the optical light curve at three x - ray flux levels .
table 5 gives the mean x - ray flux levels for the period of time between july 1 and july 16 , on august 6 and on august 15 .
the folded optical light curve for july 1 - 16 is shown in figure 3 , and figure 9 shows the v - band light curves for august 6 and august 15 . for the august observations , we obtained good , but not complete , phase coverage . for july 1 - 16 and august 6 ,
the x - ray fluxes come from rxte / asm measurements . since the fifth pointed rxte observation began immediately after our august 15 optical observation , the pca flux measurement was used . for each time interval
, we fitted the optical light curve with a sine function to determine the v - band magnitude at maximum light ( @xmath16 ) and minimum light ( @xmath17 ) . the modulation amplitude ( v(0.5)v(0.0 ) ) increases as the x - ray flux decreases .
this trend was also observed in the r - band by zurita et al .
( 1998 ) . to understand this trend , we assume that the optical flux can be separated into a modulated component coming from the optical companion and an unmodulated component coming from the accretion disk . to separate the components ,
the v - band magnitudes given in table 5 were converted into flux densities ( i.e. erg @xmath9 s@xmath10 hz@xmath10 at @xmath83 hz ) using the relationship given in table iv of bessell et al .
a flux density was also determined from the estimated quiescent v - band magnitude of 22.52 . to correct for extinction
, a@xmath84=0.3 was used . the peak contribution to the flux density from the optical companion
is given by @xmath85 , where @xmath86 is the v - band flux density at the orbital phase @xmath15 .
the contribution to the flux density from the disk is given by @xmath87 , where @xmath88 is the quiescent flux density .
the values for @xmath89 and @xmath90 are given in table 5 . from july 1 - 16 to august 15 , @xmath90 decreased by a factor of 10 , while @xmath89 only decreased by a factor of 3 . in the following , we assume that the optical light comes from x - ray reprocessing in the accretion disk and the optical companion .
first , we calculate the expected change in @xmath89 as the x - ray flux drops from its july 1 - 16 level to its august 15 level , and find that it is consistent with the observed factor of 3 .
then , a simplified model of the system is used to understand the physical changes in the system during the outburst decay .
we calculate the x - ray flux from the heated side of the optical companion by determining the temperature , @xmath91 , at each point on the star using equation 1.8 of van paradijs ( 1991 ) . assuming that each surface element emits as a blackbody at the local temperature , the integral over the surface of the star is @xmath92d\theta~,\ ] ] where @xmath93 is the ratio of the neutron star radius to the source distance . for a binary at 10 kpc consisting of a 1.4 m@xmath23 neutron star and a k8 star with an x - ray albedo of 0.4 ( @xcite ) ,
the july 1 - 16 , august 6 , and august 15 x - ray fluxes imply @xmath89 values of @xmath94 erg @xmath9 s@xmath10 hz@xmath10 , @xmath95 erg @xmath9 s@xmath10 hz@xmath10 , and @xmath96 erg @xmath9 s@xmath10 hz@xmath10 , respectively .
although these values are a factor of about 2 higher than the observed @xmath89 values , the ratio between the july 1 - 16 value and the august 15 value is about 3 , consistent with observations .
there are several reasons why our calculated @xmath89 values are higher than the observed values .
first , to simplify the calculation , we assume a binary inclination ( @xmath22 ) of 90@xmath13 . assuming that @xmath22 is not too much less than the upper limit of 73@xmath13 , this correction changes the calculated value by a factor of sin @xmath22 .
as stated above , the dip in the light curve near orbital phase 0.5 probably indicates that the accretion disk blocks part of the optical companion at maximum light , but our calculation assumes that we observe the entire optical companion unobstructed . finally ,
if the accretion disk is flared at its edge ( @xcite ) , the disk may partially shield the optical companion from the x - ray source , reducing the level of x - ray heating .
the values of @xmath90 can be used to understand the evolution of the accretion disk during the decay of the outburst .
we assume a spherical x - ray source of radius @xmath97 and a thin , azimuthally symmetric accretion disk extending radially from @xmath97 to @xmath98 and inclined at an angle @xmath22 with respect to the line of sight of the observer .
further , we assume that the v - band emission is dominated by blackbody emission from the accretion disk at a temperature @xmath91 and that the temperature is only a function of the radial coordinate , @xmath99 . based on uv measurements of other lmxbs
, we expect the accretion disk temperature to be high enough so that the v - band lies in the rayleigh - jeans portion of the blackbody spectrum ( @xcite ) . at a frequency @xmath100 ,
the flux density is @xmath101 where @xmath102 is the speed of light , @xmath103 is boltzmann s constant and @xmath104 is the distance to the source .
if the v - band emission comes from x - ray reprocessing , the temperature can be written as @xmath105 where @xmath106 is the albedo of the disk .
equation 4 assumes that @xmath97 is much less than @xmath99 .
although we integrate over values of @xmath97 to @xmath98 , this approximation is valid because most of the flux comes from close to @xmath98 . assuming that @xmath97 is much less than @xmath98 , the v - band flux density from the disk is @xmath107 where @xmath108 is the x - ray flux in units of @xmath109 erg @xmath9 s@xmath10 , @xmath110 is the radius of the x - ray source in units of @xmath111 cm , @xmath112 is the outer radius of the disk in units of @xmath113 cm and @xmath70 is the distance to the source in units of 10 kpc .
equation 5 can be used to understand how the geometry of the x - ray source and the disk change during the x - ray decline . using the values of @xmath90 and @xmath108 from table 5 ,
cos @xmath22 @xmath114 @xmath115 @xmath116 is equal to 5.8 , 2.7 , and 1.2 for july 1 - 16 , august 6 and august 15 , respectively .
only @xmath97 and @xmath98 are likely to change during the outburst decay .
@xmath98 decreases by a factor of 3.6 between july 1 - 16 and august 15 if @xmath97 is constant during the x - ray decay .
if instead we assume that the x - ray flux is proportional to the surface area of the spherical x - ray emitting region ( i.e. @xmath117 is proportional to @xmath118 ) , then @xmath98 still decreases by a factor of 2.8 between july 1 - 16 and august 15 .
we conclude that the outer radius of the accretion disk decreases as the x - ray flux decreases .
we optically identified the neutron star x - ray transient xte j2123058 , measured its orbital period and constrained the binary inclination .
the orbital period is @xmath0 hr and the binary inclination is between 55@xmath13 and 73@xmath13 .
the properties of the observed optical bursts are consistent with being reprocessed x - ray bursts .
the distance to the source is between 4.5 kpc and 15 kpc , indicating that the source is at least 2.6 kpc from the galactic plane , which is considerably greater than the 1 kpc scale height for neutron star lmxbs . to determine the distance lower limit
, we have assumed that all the quiescent optical light comes from a main sequence optical companion .
high frequency qpos at @xmath4 hz and @xmath5 hz are simultaneously detected .
according to the beat frequency model , the difference in frequency between the two qpos indicates a neutron star spin period of @xmath6 ms . during the decay of the outburst , a change in the energy spectrum
is observed , which is similar to a change observed in aql x-1 .
we speculate that this is due to the onset of the propeller mechanism , indicating that the neutron star magnetic field strength is in the range from 2 to 8 @xmath69 g. however , the changes in the timing and spectral properties observed during the decay are typical of atoll sources , which may indicate that the changes are due solely to the dynamics of the accretion disk . as the x - ray flux decays , we infer that the v - band flux density from the accretion disk drops by a factor of 10 , while the v - band flux density from the optical companion only changes by a factor of 3 . assuming the optical flux comes from x - ray reprocessing in the disk and the optical companion ,
our calculations show that the v - band flux density from the optical companion is expected to change by a factor of 3 for the observed change in x - ray flux and that the factor of 10 change in the v - band flux density from the accretion disk can be explained if the size of the accretion disk decreases significantly during the decay of the outburst .
although most of the properties of xte j2123058 are typical for neutron star lmxbs , its distance from the galactic plane is unusual . in the future
, it is important to improve constraints on its distance .
this can be accomplished from a spectroscopic determination of the spectral type of the optical companion .
also , the low galactic column density for this system provides a good opportunity to study the soft x - ray properties of a neutron star transient
. this will be especially useful for x - ray observations in quiescence ( e.g. @xcite ) .
the authors would like to thank d. hurley - keller for observing xte j2123058 , k. leighly for assisting with our observations and j. orosz for useful discussions .
p.k . and j.a.t .
acknowledge support from nasa grants nag5 - 3595 , nag5 - 3799 , nag5 - 4416 , and nag5 - 7405 .
alpar , m.a . ,
cheng , a.f . ,
ruderman , m.a .
& shaham , j. 1982 , nature 300 , 728 bessell , m.s .
1979 , pasp 91 , 589 eggleton , p.p . 1983 ,
apj 268 , 368 ford , e.c .
1997 , apj 475 , l123 homan , j. , van der klis , m. , van paradijs , j. & mendez , m. 1998 , iauc 6971 homan , j. , et al .
1999 , apj 513 , l119 hynes , r.i .
1998 , iauc 6976 illarionov , a.f . & sunyaev , r.a .
1975 , a&a 39 , 185 landolt , a.u .
1992 , aj 104 , 340 leahy , d.a .
1983 , apj 266 , 160 levine , a. , swank , j. & smith , e. 1998 , iauc 6955 lewin , w.h.g . , van paradijs , j. & taam , r.e .
1992 , space sci .
62 , 223 lewin , w.h.g . ,
van paradijs , j. & taam , r.e .
1995 , in x - ray binaries , w.h.g .
lewin , j. van paradijs & e.p.j .
van den heuvel ( eds . ) , cambridge university press , p. 175 london , r. , mccray , r. & auer , l.h . 1981 , apj 243 , 970 mendez , m. , van der klis , m. & van paradijs , j. 1998 , apj 506 , l117
monet , d. et al .
1996 , usno - sa1.0 ( washington , dc : us naval observatory ) patterson , j. 1984 , apjs 54 , 443 press , w.h . , flannery , b.p . ,
teukolsky , s.a . & vetterling , w.t .
1992 , numerical recipes : the art of scientific computing , cambridge university press psaltis , d. et al . 1998 , apj 501 , l95 rutledge , r.e .
1998 , astro - ph/9810288 schlegel , d.j .
, finkbeiner , d.p . , &
davis , m. 1998 , apj 500 , 525 stark et al .
1998 , pcabackest , http://lheawww.gsfc.nasa.gov/docs/xray/xte/pca strohmayer , t. , swank , j. & zhang , w. 1998 , astro - ph/9801219 sunyaev & titarchuk 1980 , a&a 86 , 121 takeshima , t. & strohmayer , t.e .
1998 , iauc 6958 tomsick , j.a . ,
halpern , j.p .
, leighly , k.m . & perlman , e. 1998a , iauc 6957 tomsick , j.a .
, kemp , j. , halpern , j.p . & hurley - keller , d. 1998b , iauc 6972 tomsick , j.a .
, kaaret , p. , kroeger , r.a . & remillard , r.a .
1998c , astro - ph/9809098 van der klis , m. 1995 , in x - ray binaries , w.h.g . lewin , j. van paradijs & e.p.j .
van den heuvel ( eds . ) , cambridge university press , p. 252
van der klis , m. , wijnanads , r. , horne , k. & chen , w. 1997 , apj 481 , l97 van der klis , m. 1998 , astro - ph/9812395 van paradijs , j. 1991 , in neutron stars : theory and observation , ventura , j. & pines , d. ( eds . ) , kluwer academic publishers , p. 289
van paradijs , j. & mcclintock , j.e .
1995 , in x - ray binaries , w.h.g .
lewin , j. van paradijs & e.p.j .
van den heuvel ( eds . ) , cambridge university press , p. 58
van paradijs , j. & white , n. 1995 , apj 447 , l33 white , n.e . &
holt , s.s .
1982 , apj 257 , 318 zhang , s.n . , yu , w. & zhang , w. 1998 , apj 494 , l71 zhang , w. et al .
1995 , apj 449 , 930 zurita , c. & casares , j. 1998 , iauc 7000 zurita , c. , casares , j. & hynes , r.i .
1998 , iauc 6993 cccccc 1 & june 27 & 208 & @xmath119 & june 27.98471 & june 27.986932 & june 29 & 3392 & @xmath120 & june 29.68425 & june 29.685273 & july 14 & 19808 & @xmath121 & july 14.45280 & july 14.45310 & & & & july 14.50451 & july 14.504874 & july 22 & 25888 & @xmath121 & july 22.23787 & july 22.23897 & & & & july 22.50956 & july 22.510425 & august 15 & 16064 & @xmath122 & no bursts detected & lccc june 30.332 & mdm 1.3 m & 2.74 & 30 - 300july 1.280 & mdm 2.4 m & 3.00 & 30july 1.355 & mdm 1.3 m & 0.24 & 120july 2.314 & mdm 1.3 m & 1.90 & 30july 2.396 & mdm 2.4 m & 1.83 & 30 - 60july 3.410 & mdm 2.4 m & 1.61 & 30july 4.361 & mdm 1.3 m & 2.85 & 30july 15.253 & kpno 0.9 m & 5.45 & 60july 16.255 & kpno 0.9 m & 4.70 & 60august 6.075 & ctio 0.9 m & 3.90 & 120august 15.124 & cito 0.9 m & 3.92 & 120september 20.671 & mdm 1.3 m & 1.00 & 600 lccccc comptonizationkt ( kev ) & @xmath123 & @xmath124 & @xmath125 & @xmath126 & @xmath127@xmath61 & @xmath128 & @xmath129 & @xmath130 & @xmath131 & @xmath132flux & 206 & 173 & 99 & 100 & 7.6n@xmath133 ( @xmath134
@xmath9 ) & @xmath135 & @xmath136 & @xmath137 & @xmath138 & @xmath139@xmath59 & 40/44 & 46/44 & 65/44 & 52/44 & 36/44comptonizationkt ( kev ) & @xmath140 & @xmath141 & @xmath142 & @xmath143 & @xmath144@xmath61 & @xmath145 & @xmath146 & @xmath147 & @xmath148 & @xmath149flux & 194 & 145 & 95 & 95 & 7.6@xmath59 & 48/45 & 77/45 & 97/45 & 92/45 & 37/45blackbody plus power - lawkt ( kev ) & @xmath150 & @xmath151 & @xmath152 & @xmath64 & @xmath65r@xmath153 ( in km for @xmath104=10 kpc ) & @xmath154 & @xmath155 & @xmath156 & @xmath157 & @xmath158@xmath58 & @xmath159 & @xmath160 & @xmath161 & @xmath66 & @xmath67flux@xmath162 & 147 & 123 & 77 & 75 & 7.4n@xmath133 ( @xmath134 @xmath9 ) & @xmath163 & @xmath164 & @xmath165 & @xmath166 & @xmath167@xmath59 & 35/43 & 44/43 & 60/43 & 45/43 & 26/43blackbody plus power - lawkt ( kev ) & @xmath168 & @xmath169 & @xmath170 & @xmath171 & @xmath172r@xmath153 ( in km for @xmath104=10 kpc ) & @xmath173 & @xmath174 & @xmath175 & @xmath176 & @xmath177@xmath58 & @xmath178 & @xmath179 & @xmath180 & @xmath181 & @xmath182flux@xmath162 & 131 & 96 & 62 & 58 & 7.4@xmath59 & 38/44 & 103/44 & 154/44 & 149/44 & 26/44 ccccccc observation & n & @xmath73 & rms amplitude & @xmath75 & @xmath76 ( hz ) & rms amplitude2 & 15 & @xmath183 & @xmath184 % & 0.6 & 27 & @xmath185 % 3 & 143 & @xmath186 & @xmath187 % & @xmath188 & @xmath189 & @xmath77 % 4 & 183 & @xmath190 & @xmath191 % & @xmath192 & @xmath193 & @xmath78 % 5 & 117 & @xmath194 & @xmath79 % & 0.5 & 7 & @xmath195 % | we report on optical and rxte observations of a new high - latitude bursting x - ray transient , xte j2123058 .
we identified the optical counterpart and discovered a @xmath0 hr periodic optical modulation , which was subsequently shown to be the same as the spectroscopic orbital period . from the absence of orbital x - ray modulation and the presence of partial
optical eclipses the binary inclination is between @xmath1 and @xmath2 . from the optical magnitude in quiescence and from the x - ray flux of type i x - ray bursts ,
the source distance is between 4.5 and 15 kpc , which implies that the source is unusually far from the galactic plane since @xmath3 .
optical bursts with properties consistent with being reprocessed x - ray bursts occurred .
we detect a pair of high frequency qpos at @xmath4 hz and @xmath5 hz simultaneously . according to the beat frequency model
, this qpo separation implies a neutron star spin period of @xmath6 ms . a change in the energy spectrum occurred during the decay of the outburst , which may have been due to the onset of the propeller mechanism .
if so , then the neutron star magnetic field strength is between 2 and 8 @xmath7 g for an assumed distance of 10 kpc .
however , the changes in the timing and spectral properties observed during the decay are typical of atoll sources , which may indicate that the changes are due solely to the dynamics of the accretion disk .
as the phase averaged v - band magnitude declined from 17.26 at the peak of the outburst to 19.24 , and the x - ray flux decreased from @xmath8 erg @xmath9 s@xmath10 to @xmath11 erg @xmath9 s@xmath10 , the peak - to - peak amplitude of the v - band modulation increased from 0.75 to 1.49 magnitudes .
this behavior can be explained if the size of the accretion disk decreases during the decay of the outburst . |
recently , a long - standing problem or puzzle @xcite in any relativistic quantum field theories has been investigated by the present authors @xcite .
the problem is how to construct the negative energy sea , or dirac sea for bosons , since as is well known the fermion fields was historically second quantized firstly by dirac in terms of dirac sea and hole theory @xcite . in the fermion case there
exists pauli s exclusion principle and easily negative energy sea , namely , the dirac sea is constructed . in the bosonic cases contrary to fermions , one might think at first that it would be impossible to construct such a sea due to lack of the pauli principle , so that infinite number of bosons at each energy state could exist and thus the negative energy states could never be filled .
however , we succeeded in constructing the dirac sea for bosons , so called _ boson sea_. in fact , there we solved one of the serious problems : how to construct the positive definite norm of the negative energy states .
there we have used a non - local definition ( the detail of the methods see @xcite ) .
it is the purpose of the present article to show another method employing the regularization of the naively divergent inner product in the negative number sector and the renormalization .
in fact we make use of a kind of @xmath0-regularization method to make it finite , and then make renormalization by discarding all the divergent terms , which can be done successfully .
the advantage of this @xmath0-method is to make transparent the analytic structure of the whole procedure .
the present paper is organized as follows : in the following section 2 we treat the inner product by @xmath0-regularization and renormalization , and the positive definite inner product is obtained . in section 3
we verify the orthonormality of our inner product obtained in section 2 by performing some explicit analytic calculation . in section 4 we present another definition of the inner product without the subtraction scheme in the renormalization , and we give a proof of the orthonormality .
section 5 is devoted to conclusion and further perspectives .
as a preparation to define the inner product , we define an @xmath9-regularized inner product as @xmath10 where @xmath11 is the integral region , and @xmath12 is a regularization function .
the integral region @xmath13 is just the inside of the light - cone shown as the shaded zone in fig .
[ gamma ] . on @xmath14-plane .
the exponential factor @xmath15 converges in this region . ] the @xmath9-regularized inner product ( [ pre - gamma - product ] ) is divergent for @xmath16 , and it is divided into the following three parts : @xmath17 by the behavior of @xmath16 . the first term @xmath18 diverges for @xmath16 .
according to the precise calculation presented in the next section , the concrete form of @xmath18 is given by the linear combination of @xmath19 for positive integer n. thus we can manifestly separate the second term @xmath20 which is just independent term of @xmath9 for @xmath16 .
the third term @xmath21 goes to zero for @xmath16 .
we define the inner product by a renormalization of the @xmath9-regularized inner product : @xmath22 which may be so called the minimal subtraction scheme of the renormalization .
we can confirm the inner product ( [ gamma - product ] ) satisfies the orthonormal condition : @xmath23 the product ( [ gamma - product ] ) is just positive definite even for the indefinite metric of @xmath24-algebra , namely , @xmath25 .
therefore we construct the hilbert space including the negative number sector by using the product ( [ gamma - product ] ) .
these are the result from the combination of the @xmath13-restriction , the regularization function @xmath26 and the renormalization . in the definition of the inner product ( [ gamma - product ] ) , the restriction of the integral region into @xmath13 and the regularization function @xmath26 are quite important .
the restriction of the integral region into @xmath13 is a part of the regularization which is nothing but a kind of hard cut - off .
the choice of the integral region @xmath13 is important to realize the orthogonal condition .
we hereby verify the orthonormality ( [ ortho - normal - condition ] ) of the inner product ( [ gamma - product ] ) . we introduce a hyperbolic coordinate @xmath27 on the @xmath14-plane for convenience .
the hyperbolic coordinate @xmath27 on the @xmath14-plane is defined as @xmath28 where @xmath29 and @xmath30 covers the whole region @xmath13 as shown in fig .
[ hyperbola ] . for fixed @xmath31 .
] this hyperbolic coordinate respects the lorentz invariance of the hamiltonian and the `` gaussian '' factor @xmath15 ( see our previous paper @xcite ) .
the relation between the differential operators is given by @xmath32 \frac{\partial}{\partial y } \end{array } \right ) & = & \left ( \begin{array}{cc } + \cosh\theta & -\frac{1}{r}\sinh\theta\\[0.5em ] -\sinh\theta & + \frac{1}{r}\cosh\theta \end{array } \right ) \left ( \begin{array}{c } \frac{\partial}{\partial r}\\[0.5em ] \frac{\partial}{\partial\theta } \end{array } \right),\end{aligned}\ ] ] and the relation between integral measures becomes @xmath33 as the first step of the proof , we concretely calculate several inner products and norms of the states . the most important is the vacuum norm .
the @xmath9-regularized product ( [ pre - gamma - product ] ) of the vacuum becomes @xmath34 then we obtain @xmath35 , because there arises no divergent part in ( [ pre - vacuum - norm ] ) and the @xmath9-independent part of ( [ pre - vacuum - norm ] ) is @xmath36 as @xmath37 .
we also calculate a product of the orthogonal states @xmath38 in this case , the orthogonal relation is realized without any regularization .
one of the non - trivial cases is @xmath39 thus the renormalized product defined in ( [ gamma - product ] ) becomes @xmath40 , because there appears no @xmath9-independent term rather than the divergent term .
more important case is the following : @xmath41 where we have used the calculation steps ( [ pre - vacuum - norm ] ) and ( [ pre-02-norm ] ) .
we obtain @xmath42 .
we have the positive value of this product even for the negative norm of the @xmath43-element . as the second step of the proof , we derive recurrence formulae for the @xmath9-regularized inner product . for any wave - functions @xmath44 and @xmath45 , we have the following relations : @xmath46_\varepsilon \;+\ ; 2 \varepsilon \langle { f } \:| x |\ : { g } \rangle_\varepsilon , \label{dif - x } \\
\bigl\langle f \:\bigl|\ : { \textstyle\frac{d}{dy}g } \bigr\rangle_\varepsilon & = & \;-\ ; \bigl\langle { \textstyle\frac{d}{dy}f } \:\bigl|\ : { g } \bigr\rangle_\varepsilon \;+\ ; { { \cal s}}_y\left [ \left < f , g\right > \right]_\varepsilon , \label{dif - y}\end{aligned}\ ] ] where we have defined surface terms as @xmath47_\varepsilon & = & + \frac{1}{-\log\varepsilon } \left [ \int_{-\infty}^{+\infty } d\theta \ :
|r| \cosh\theta \cdot \left < f , g\right > \cdot e^{-\epsilon r^2 \cosh^2 \theta } \right]_{r=-\infty}^{r=+\infty } \nonumber\\ & & - \frac{1}{-\log\varepsilon } \left [ \int_{-\infty}^{+\infty } dr \ : |r| \frac{\sinh\theta}{r } \cdot \left
< f , g\right > \cdot e^{-\epsilon r^2 \cosh^2 \theta } \right]_{\theta = -\infty}^{\theta = + \infty } , \\ { { \cal s}}_y\left [ \left < f , g\right >
\right]_\varepsilon & = & - \frac{1}{-\log\varepsilon } \left [ \int_{-\infty}^{+\infty } d\theta \ : |r| \sinh\theta \cdot \left
< f , g\right > \cdot e^{-\epsilon r^2 \cosh^2 \theta } \right]_{r=-\infty}^{r=+\infty } \nonumber\\ & & + \frac{1}{-\log\varepsilon } \left [ \int_{-\infty}^{+\infty } dr \ : |r| \frac{\cosh\theta}{r } \cdot \left < f , g\right > \cdot e^{-\epsilon r^2 \cosh^2 \theta } \right]_{\theta = -\infty}^{\theta = + \infty}.\end{aligned}\ ] ] we postpone presenting details of the derivation ( [ dif - x ] ) to appendix .
the third term in ( [ dif - x ] ) comes from the regularization function @xmath26 in the definition of the @xmath9-regularized product ( [ pre - gamma - product ] ) . by applying the relation ( [ dif - x ] ) into the creation operator @xmath48 of the positive number sector
, we obtain the following relation : @xmath49_\varepsilon \nonumber\\ & & \;-\ ; \sqrt{2 } \varepsilon \langle \phi_{n_+,-m_- } \:|\:x\ : \phi_{n'_+-1,-m'_- } \rangle_\varepsilon .
\label{aa - x}\end{aligned}\ ] ] by using the relation @xmath50 in the third term in ( [ aa - x ] ) and by operating the creation and annihilation operators on the wave functions , the relation ( [ aa - x ] ) becomes a relation among three energy levels : @xmath51_\varepsilon \nonumber\\ & & \;-\ ; \sqrt{n'-1 } \varepsilon \langle \phi_{n_+,-m_- } \:| \ :
\phi_{n'_+-2,-m'_- } \rangle_\varepsilon .
\label{level - x}\end{aligned}\ ] ] we can derive the similar relation to ( [ aa - x ] ) and ( [ level - x ] ) for the negative number sector .
this derivation is simpler than that of the positive number sector , due to the absence of the third term in ( [ dif - y ] ) .
in fact by using the relation ( [ dif - y ] ) in the annihilation operator @xmath52 , we obtain @xmath53_\varepsilon , \label{aa - y}\end{aligned}\ ] ] where we have used the property @xmath54 . by the operation of the creation and annihilation operators , the relation ( [ aa - y ] ) becomes @xmath55_\varepsilon .
\label{level - y}\end{aligned}\ ] ] the equations ( [ level - x ] ) and ( [ level - y ] ) are recurrence formulae which determine the concrete values of the @xmath9-regularized inner product .
the first terms of the recurrence formulae are given by the @xmath9-regularized norm of vacuum in ( [ pre - vacuum - norm ] ) , namely , @xmath56 and the properties @xmath57 we consider @xmath9-behavior of the surface term in the recurrence formula ( [ level - x ] ) .
the surface term for any @xmath58 and @xmath59 is given by linear combination : @xmath60_\varepsilon & = & { \sum_{a=0}^{n_++n_+'-1 } \ ; \sum_{b=0}^{m_-+m_- ' } } c_{a , b } \ : { { \cal s}}_x \left [ x^a y^b e^{-x^2 + y^2 } \right]_\varepsilon , \label{linear - combination - x}\end{aligned}\ ] ] where @xmath61 is a coefficient of the linear combination , because @xmath62_\varepsilon$ ] is a linear functional and the product @xmath63 is obtained by the linear combination of the functions ; @xmath64 for integers
@xmath65 . in the last term of ( [ linear - combination - x ] ) reads @xmath66_\varepsilon & = & \frac{-2}{-\log\varepsilon } \ : \left(\frac{1}{\varepsilon}\right)^{\frac{a+b+1}{2 } } \ : \times \left\ { \begin{array}{ll } 1 & ( a=1,\ b=0)\\ \frac{a+b-1}{2}\gamma\left(\frac{a+b-1}{2}\right ) & ( a:{\rm odd},\ b:{\rm even})\\ 0 & ( { \rm others } ) \end{array } \right.,\end{aligned}\ ] ] which is zero or diverges for @xmath67 .
the left - hand side of ( [ linear - combination - x ] ) consists of @xmath9-divergent terms and contains no @xmath9-independent terms , so that we conclude that the surface term in ( [ level - x ] ) never contribute to the renormalized inner product defined in ( [ gamma - product ] ) .
the surface term in ( [ level - y ] ) has the same @xmath9-behavior as that in ( [ level - x ] ) , because the surface term @xmath68_\varepsilon $ ] is obtained by a linear combination of the functions @xmath64 , and we have a relation @xmath69_\varepsilon & = & \frac{2}{-\log\varepsilon } \ : \left(\frac{1}{\varepsilon}\right)^{\frac{a+b+1}{2 } } \ : \times \left\ { \begin{array}{ll } 1 & ( a=0,\ b=1)\\ \frac{a+b-1}{2}\gamma\left(\frac{a+b-1}{2}\right ) & ( a:{\rm even},\ b:{\rm odd})\\ 0 & ( { \rm others } ) \end{array } \right .. \end{aligned}\ ] ] thus the surface term in ( [ level - y ] ) has no contribution to the renormalized inner product ( [ gamma - product ] ) .
as we have seen above , the surface terms in the recurrence formulae ( [ level - x ] ) and ( [ level - y ] ) consist of divergent terms whose @xmath9-dependence is @xmath70 when we multiply ( [ epsilon - dependence ] ) by @xmath9 , the product never contain @xmath9-independent part . thus the third term in ( [ level - x ] )
never contribute to the renormalized inner product ( [ gamma - product ] ) .
finally the recurrence formulae ( [ level - x ] ) and ( [ level - y ] ) tell us that @xmath9-regularized inner product ( [ pre - gamma - product ] ) has the property : @xmath71 and we conclude that the renormalized inner product ( [ gamma - product ] ) satisfies the orthonormal condition : @xmath72
one may consider that the separation of the divergent part in ( [ separation ] ) has ambiguity and the definition of the renormalization ( [ gamma - product ] ) is also ambiguous .
as we have seen in the proof in the section 3 , there is no ambiguity because the structure of the divergent part is completely understood . in this section ,
we present another definition of the inner product which obviously has no ambiguity because the subtraction scheme is not employed .
we extend the regularization parameter @xmath9 in ( [ pre - gamma - product ] ) from real value to complex one .
the definition of the inner product is @xmath73 , \label{holomorphic - product}\end{aligned}\ ] ] where we have introduced the parameterization of @xmath9 as @xmath74 namely , the positive parameter @xmath75 is the absolute value , and the real parameter @xmath76 is the phase .
the limit of the phase @xmath76 in ( [ holomorphic - product ] ) is continuously taken according to the path @xmath77 the absolute value @xmath75 should be always smaller than 1 in the limit .
the order of the limit in ( [ holomorphic - product ] ) is quite important , namely , the limit of @xmath76 should be taken before the limit of @xmath75 .
there is no subtraction in the definition ( [ holomorphic - product ] ) , thus there never arises ambiguity in the inner product .
we will proof the orthonormality of the inner product ( [ holomorphic - product ] ) .
first we confirm that the norm of the vacuum @xmath78 becomes unity . by recalling the relation ( [ pre - vacuum - norm ] ) ,
the vacuum norm becomes @xmath79 \nonumber\\ & = & \lim_{\alpha \to 0 } \left [ \lim_{\varphi \to \infty } \frac{1}{\sqrt{\varepsilon + 1 } } \times \frac { \log\left(\sqrt{\varepsilon + 1}+1\right ) -\log\left(\sqrt{\varepsilon + 1}-1\right)}{-\log\varepsilon } \right ] .
\label{vacuum - norm-2a}\end{aligned}\ ] ] by taking the path ( [ phi_path ] ) , the first factor @xmath80 in ( [ vacuum - norm-2a ] ) has no extraordinary contribution , because @xmath81 and the path ( [ phi_path ] ) does not cross the branch - cut of the square root . in the second factor of ( [ vacuum - norm-2a ] ) , while the pass never crosses the branch cut of @xmath82 in the numerator , the pass go across the branch cut of @xmath83 in the denominator and of @xmath84 in the numerator .
therefor we find that the vacuum norm ( [ vacuum - norm-2a ] ) becomes unity as @xmath85 \nonumber \\ & = & 1 .
\label{vacuum - norm-2}\end{aligned}\ ] ] according to the argument in the previous section , the ( @xmath9-divergent part ) in ( [ resultant - separation ] ) consists of the terms in ( [ epsilon - dependence ] ) , and the ( @xmath9-zero pert ) consists of the terms @xmath86 by taking the limit according to the path ( [ phi_path ] ) , the factor @xmath87 in the ( @xmath9-divergent part ) and in the ( @xmath9-zero pert ) has no additional contribution because @xmath88 is a integer , and the only logarithm in the denominator produces the additional phase factor @xmath89 .
thus , these terms behave in the limit as @xmath90 & = & \lim_{\alpha \to 0 } \left [ \lim_{\varphi \to
\infty } \frac{1}{-(\log\alpha ) - i \varphi } \right ] \alpha^{\pm n } \nonumber\\ & = & 0,\end{aligned}\ ] ] where @xmath88 is a positive integer .
we have found that the ( @xmath9-divergent part ) and the ( @xmath9-zero pert ) vanish in the limit of @xmath76 .
therefore , the inner product defined in ( [ holomorphic - product ] ) results the orthonormal relation : @xmath91 \nonumber\\ & = & \delta_{n_+,n_+ ' } \delta_{m_-,m_-'}.\end{aligned}\ ] ]
we have proposed the other definitions of the positive definite inner product for the negative number sector of the double harmonic oscillator than the non - local method in the previous paper @xcite .
the new definitions provide positive definite inner product , and there arises no difference of results among the definitions . the separation of @xmath20 from @xmath18 in ( [ separation ] ) has been succeeded , because we have known the form of @xmath18 by the concrete calculation .
one may consider that the separation in ( [ separation ] ) may seem to be ambiguous , however , we have found that the anxiety of the ambiguity is completely solved by the holomorphic definition ( [ holomorphic - product ] ) .
this means that there arises no ambiguity in the perturbation theories , because the perturbation theories never deform the harmonic oscillators in the mode expansion of the fields .
when we consider the non - perturbative phenomena , i.g .
, quantum anomalies in external fields , the ambiguity may arises and results the non - trivial effects .
this is the further subject to investigate .
we acknowledge r. jackiw for his encouraging communication who shares with us the view point that the dirac sea method provides deep physical understanding and intuition to some novel phenomena .
we also thank t. iwaki , z. tsuboi and t. tsuchida for useful discussions .
this work is supported by grants - in - aid for scientific research on priority areas , number of areas 763 , `` dynamics of strings and fields '' , from the ministry of education of culture , sports , science and technology , japan .
we present a detailed derivation of ( [ dif - x ] ) in the following : @xmath92_{r=-\infty}^{r=+\infty } \nonumber\\ & & \qquad - \left [ \int_{-\infty}^{+\infty } dr \ : \frac{|r|}{r } \sinh\theta \cdot fg \cdot e^{-\epsilon r^2 \cosh^2 \theta } \right]_{\theta = -\infty}^{\theta = + \infty } \nonumber\\ & & \qquad -\int_{-\infty}^{+\infty } dr \int_{-\infty}^{+\infty } d\theta \ : \left [ \left\{\frac{d}{dr } \left(|r| f\right ) \right\ } \cosh\theta \cdot g -\frac{|r|}{r}\left\{\frac{d}{d\theta } \left(f\sinh\theta\right)\right\}g \right ] \cdot e^{-\epsilon r^2 \cosh^2 \theta } \nonumber \\ & & \qquad -\int_{-\infty}^{+\infty } dr \int_{-\infty}^{+\infty } d\theta \ : |r| f g \left\ { \cosh\theta \cdot \frac{d}{dr } -\frac{\sinh\theta}{r } \frac{d}{d\theta } \right\ } e^{-\epsilon r^2 \cosh^2 \theta } \nonumber \\ & = & [ \mbox{surface terms } ] \nonumber \\ & & \qquad -\int_{-\infty}^{+\infty}dr \int_{-\infty}^{+\infty}d\theta \ : |r| \left[\left\{\cosh \theta \frac{d}{dr}f\right\ } -\left\{\frac{\sinh \theta}{r}\frac{d}{d\theta}f\right\ } \right ] \cdot g \cdot e^{-\epsilon r^2 \cosh^2 \theta } \nonumber \\ & & \qquad -\int_{-\infty}^{+\infty } dr \int_{-\infty}^{+\infty } d\theta \ : |r| f g \left(-2 \varepsilon \right ) r \cosh \theta \ : e^{-\epsilon r^2 \cosh^2 \theta } \nonumber \\ & = & [ \mbox{surface terms } ] \;-\ ; \int_{\gamma } dxdy \ : \frac{d}{dx } f \cdot g \cdot e^{-\epsilon x^2 } \;+\ ; 2 \varepsilon \int_{\gamma } dxdy \ : f \cdot g \ : x \cdot e^{-\epsilon x^2 } \nonumber\\ & = & [ \mbox{surface terms } ] \;-\ ; \left < { \textstyle\frac{d}{dx}f } \:|\ : { g } \right>_\varepsilon \;+\ ; 2 \varepsilon \left < { f } \:| x |\ : { g } \right>_\varepsilon .
\label{h.2.12}\end{aligned}\ ] ] 99 s. weinberg , `` the quantum theory of fields .
vol . 1 : foundations , '' cambridge university press ( 1995 ) . h. b. nielsen and m. ninomiya , arxiv : hep - th/9808108 . y. habara , h. b. nielsen and m. ninomiya , int
. j. mod .
a * 19 * , 5561 ( 2004 ) [ arxiv : hep - th/0312302 ] .
h. b. nielsen and m. ninomiya , prog .
. phys . * 113 * , 603 ( 2005 ) [ arxiv : hep - th/0410216 ] ; prog .
phys . * 113 * , 625 ( 2005 ) [ arxiv : hep - th/0410218 ] .
y. habara , y. nagatani , h. b. nielsen and m. ninomiya , arxiv : hep - th/0603242 .
p. a. m. dirac , proc .
a * 126 * , 360 ( 1930 ) ; proc .
* 26 * , 361 ( 1930 ) . | in bosonic formulation of the negative energy sea , so called dirac sea presented in the preceding paper [ arxiv : hep - th/0603242 ] , one of the crucial points is how to construct a positive definite inner product in the negative energy states , since naive attempts would lead to non - positive definite ones . in the preceding paper
the non - local method is used to define the positive definite inner product . in the present article
we make use of a kind of @xmath0-regularization and renormalization method which may clarify transparently the analytical properties of our formulation .
yitp-06 - 35 + kuns-2032 + oiqp-06 - 09 + hep - th/0607182 * dirac sea and hole theory for bosons ii * + _ renormalization approach _ + yoshinobu habara@xmath1 , yukinori nagatani@xmath2 , + holger b. nielsen@xmath3 and masao ninomiya@xmath4 + _ @xmath5department of physics , graduate school of science , + kyoto university , kyoto 606 - 8502 , japan + @xmath6okayama institute for quantum physics , + kyoyama - cho 1 - 9 , okayama - city 700 - 0015 , japan + @xmath7niels bohr institute , university of copenhagen , + 17 blegdamsvej copenhagen , denmark + @xmath8yukawa institute for theoretical physics , + kyoto university , kyoto 606 - 8502 , japan _ |
one of the main subjects of current interest in condensed matter physics is the search for materials that host topological insulator ( ti ) phases @xcite .
two dimensional tis exhibit the quantum spin hall effect ( qshe ) with gapless edge states and a finite energy gap in the bulk @xcite .
the first proposal of this state of matter was made by kane and mele @xcite on the basis of graphene in the presence of spin - orbit interaction ( soi ) .
however , the relevant soi in graphene turns out to be rather small @xcite such that the effect seems to be inaccessible in experiments .
this situation is different in hgte / cdte quantum wells where the qshe was also predicted theoretically @xcite and experimentally seen soon after @xcite .
recently , a single layer of silicon atoms called silicene has been synthesized exhibiting an analogous honeycomb structure as graphene @xcite .
since silicon is heavier than carbon , the spin - orbit gap in silicene is much larger than in graphene .
therefore , if it was possible at some point to prepare clean silicene , it should be feasible to experimentally access the qshe in this material . similar to graphene ,
the unit cell of silicene contains two atoms which gives rise to two different sublattices .
in contrast to graphene , however , the silicene sublattices are found to be arranged in a buckled structure pointing out - of - plane @xcite . due to the broken sublattice symmetry ,
the mobile electrons in silicene are therefore able to couple differently to an external electric field than the ones in graphene .
this difference is the origin of new ( rashba ) spin - orbit coupling effects that allow for external tuning and closing of the band gap in silicene @xcite .
consequently , an electrically induced topological quantum phase transition is possible .
it is natural to ask whether this phase transition can in principle go both ways , i.e. , whether the electric field can be used to destroy and generate the qshe . refs . @xcite
clearly show that a different potential on the two sublattices of a honeycomb lattice leads to a transition from a ti to a trivial insulating state . in ref .
@xcite , some indications have been presented that the interplay of two silicene specific ( rashba ) spin - orbit terms can even induce the qshe starting from a trivial insulating band structure in the absence of these terms . the quantum spin hall ( qsh ) phase is distinguished from a normal insulating phase by a bulk @xmath1 topological invariant @xcite . for a minimal model of the qshe in graphene ,
this invariant has been analytically calculated in a seminal work by kane and mele @xcite .
however , the original formulation of the @xmath1 invariant in terms of bloch functions does not contain a constructive prescription as to its numerical evaluation .
subsequent work on the topological properties of the band structure of silicene was restricted to the absence of terms breaking the spin @xmath2-conservation or to employing the bulk - boundary correspondence in silicene nanoribbons @xcite .
evidently , a full topological analysis of silicene is missing and , as we show below , important to clearly identify phenomenological differences between graphene and silicene . in this work ,
we employ prodan s method @xcite to calculate the topological invariant without any further symmetry assumptions in a manifestly gauge invariant way to provide a conclusive analysis of the novel features of silicene regarding qsh physics .
in particular , we establish that , in contrast to graphene , the qshe can be generated by ( rashba ) soi in silicene .
the bulk of this article consists of two parts which we keep fairly self contained to allow the reader to follow our analysis la carte . in sec . 2
, we analyze the symmetries of the lattice of silicene .
this analysis allows us to mathematically construct the low - energy hamiltonian ( close to the @xmath0 points of the brillouin zone ) by means of the invariant expansion method with a particular focus on terms involving a perpendicular electric field .
thereby , we discover for silicene an additional , electric field induced spin - orbit term of the low energy hamiltonian .
furthermore , a tight - binding calculation is performed to verify the terms previously derived from symmetries and to estimate their magnitude .
the reader who is more interested in quantum spin hall physics can directly go to sec . 3 , where we study the topological properties of the band structure of silicene by explicitly calculating the @xmath1 topological invariant in a manifestly gauge invariant way .
in this section , the possibility of a topological phase transition induced by an external field is carefully examined which enables us to correct previously proposed phase diagrams . finally , we conclude in sec .
some technical details of the invariant expansion and the tight - binding model are presented in the appendix .
silicene is a monolayer of silicon atoms arranged in a buckled honeycomb lattice ( see fig .
[ fig : silicene ] for a schematic ) .
in contrast to graphene , the two basis atoms of the unit cell ( called @xmath3 and @xmath4 ) are separated perpendicular to the atomic plane at a distance @xmath5 with @xmath6 @xcite . as there is no translation symmetry in the out - of - plane direction ,
the material is quasi - two - dimensional . the buckling is quantified by an angle @xmath7 as shown in fig .
[ fig : silicene ] .
schematic of the real lattice of silicene .
the sublattices ( denoted @xmath3 and @xmath4 ) of the honeycomb structure are spatially separated in @xmath8-direction .
the buckling - angle @xmath9 is found to be @xmath10 in a silicene lattice model @xcite . ] in fig .
[ fig : geo ] , we provide an illustration of the lattice of silicene in real and reciprocal space .
basis vectors defining the unit cell are given by @xmath11 in real space and by @xmath12 in reciprocal space .
here , @xmath13 is the distance between two neighboring silicon atoms .
symmetry operations , marked as dashed and dotted lines , of the reciprocal ( a ) and the real lattice ( b ) .
the operations @xmath14 denote rotations by @xmath15 around the @xmath8-axis perpendicular to the plane ; @xmath16 and @xmath17 refer to rotations around the labeled axis , lying within the atomic plane .
@xmath18 describes reflection planes spanned by the labeled axis and the @xmath8-axis .
the primed operations cross the corners of the underlying hexagon , while the double - primed ones do not .
@xmath3 and @xmath4 denote different sublattices , @xmath0 and @xmath19 inequivalent corner points of the first brillouin zone .
the red dots implicate , that for silicene in real space the atomic sites are shifted perpendicular to the plane . ] in real space without buckling , the lattice exhibits @xmath20 symmetry ( i.e. the graphene case ) .
all the symmetry operations sketched in fig .
[ fig : geo ] are present , as well as bulk inversion @xmath21 and reflection at the atomic plane itself , @xmath22 . with the buckling present ( silicene ) ,
symmetries @xmath23 , @xmath24 , @xmath25 , @xmath26 and @xmath22 are broken and we are left with point group @xmath27 .
+ let us now go to reciprocal space . for symmorphic groups , the @xmath28 point always has the same symmetry as the real lattice .
however , the notation in use of symmetry axes with a single prime referred to the fact , that these axes cross the corners of the underlying hexagon ( see the caption of fig . [
fig : geo ] ) . under fourier transformation ,
the hexagonal lattice is rotated by @xmath29 with respect to the axes of a fixed coordinate system .
the positions of the @xmath24- and @xmath18-axes stay the same then , while the corners of the hexagon come to rest on the previously double - primed symmetry axes now . to have a corresponding group theoretical notation in reciprocal space as well , we rename the symmetry axes .
fourier transformation of the lattice can thus effectively be considered as an interchange of single and double - primed operations ( see fig .
[ fig : geo ] ) . in particular , the @xmath27-symmetry at the @xmath28-point is equivalent to operations @xmath30 , @xmath25 , @xmath26 , @xmath21 and @xmath31 in reciprocal space , which corresponds to the point group @xmath32 with the additional symmetry classes @xmath33 . at the @xmath0 points of the brillouin zone
, the symmetry of the group of the wave vector is further reduced to the point group @xmath32 .
the full knowledge of the lattice symmetries makes it in principle possible to construct a low - energy hamiltonian by expansion around high - symmetry points .
this can be done with powerful and well - established approaches , for instance , the invariant expansion @xcite . in undoped silicene
, the fermi level is located at the @xmath0-points of the brillouin zone .
hence , a low - energy hamiltonian constructed by a symmetry analysis near the @xmath0-points will capture the essential physical properties of the system .
+ we perform an invariant expansion around the @xmath0-points of silicene , which were identified to exhibit the symmetry point group @xmath32 .
the @xmath34-orbital wavefunction transforms like the two - dimensional ir @xmath35 of the group @xmath32 , while the spin part is represented by the ir @xmath36 of the double group .
therefore , our total wavefunction is of the form of the product @xmath37 and the starting point for the derivation of the hamiltonian .
a detailed presentation of this expansion is given in [ sec : appa ] .
+ consequently , the low - energy hamiltonian of silicene near the @xmath0-points is found to be ( in lowest orders of @xmath38 and @xmath39 ) @xmath40 \nonumber+\\ & a_{11 } e_z \sigma_0 ( s_x k_y - s_y k_x ) + a_{13}\sigma_z ( s_x k_y - s_y k_x ) + \nonumber \\ & a_{16}e_z(\sigma_x k_x + \tau_z\sigma_y k_y)s_z , \nonumber\end{aligned}\ ] ] presented in the basis @xmath41 , where @xmath42 , @xmath43 . here , @xmath44 distinguishes the inequivalent valleys @xmath0 and @xmath19 .
the pauli matrices @xmath18 act on sublattices @xmath3 and @xmath4 , while @xmath45 are the corresponding matrices in spin space .
additional constraints due to time - reversal symmetry ( trs ) have already been taken into account , as explained in [ sec : appa ] .
+ in eq .
( [ eq : ham2 ] ) , the terms proportional to @xmath46 , @xmath47 , and @xmath48 are well known from the graphene literature to describe its fundamental properties near the @xmath0-points @xcite .
further corrections proportional to @xmath49 and @xmath50 can as well be found in graphene @xcite . beyond this , silicene exhibits specific terms proportional to @xmath51 and @xmath52 as reported in refs .
@xcite . additionally , we find an electric field induced soi - term proportional to @xmath53 , that has not been reported for silicene before .
moreover , the hamiltonian exhibits higher order corrections to the linear dispersion ( proportional to @xmath54 ) .
we conclude , that the low - energy hamiltonian of silicene near the @xmath0-points contains interesting soi terms that are absent in graphene because of the difference in the symmetry of the two honeycomb lattices . to supplement the symmetry analysis , a corresponding tight - binding model is discussed next .
this model enables us to construct a valid hamiltonian for the full brillouin zone which is crucial for the subsequent topological analysis .
let us start with a brief discussion of the internal spin - orbit coupling and subsequently introduce the other important terms of the tight - binding model . in real space , silicene is described in terms of a lattice model including the ( standard ) spin - orbit coupling term @xcite @xmath55 where @xmath56 is the force stemming from the electric potential @xmath57 , @xmath58 is the momentum , and @xmath59 the spin of the electron .
the ( internal ) electric force in silicene is provided by the crystal field in in - plane and out - of - plane direction .
since the momentum operator is oriented along nearest neighbor or next - nearest neighbor bonds , the silicene lattice forbids terms involving a crystal in - plane force coupling to a nearest neighbor bond .
this consideration leads to the following spin - orbit hamiltonian @xcite @xmath60 with , so far , undefined parameters @xmath61 and @xmath62 . in the latter equation , @xmath63 and @xmath64
are spin quantum numbers ; the indexes @xmath65 label the atomic site / orbital .
here and in the following , @xmath66 denotes nearest neighbors and @xmath67 next - nearest neighbors .
the neighboring sites are each time connected by the vector @xmath68 with its corresponding unit vector @xmath69 .
the sign @xmath70 refers to the next - nearest neighbor hopping being anticlockwise or clockwise with respect to the positive @xmath8-axis .
furthermore , @xmath71 is introduced to distinguish between the @xmath72 site .
additionally , there is a regular nearest - neighbor hopping term and on - site energies of the form @xmath73 where @xmath74 is the on - site energy of the atomic orbital and @xmath75 the hopping parameter for hopping between the orbitals @xmath21 and @xmath76 of neighboring atomic sites .
when an external electric field @xmath39 is applied perpendicularly to the atomic plane of silicene , a staggered sublattice potential of the form @xmath77 is generated @xcite with a , so far , undefined parameter @xmath78 .
interestingly , @xmath79 allows for on - site transitions between the @xmath80 and @xmath45 orbitals @xcite . to describe a full next - nearest neighbor tight - binding model
, we also introduce spin - orbit terms involving external electric forces in eq .
( [ eq : hatomso ] ) , which we may index as @xmath81 here due to their resemblance to rashba terms .
the simplest ones are @xmath82 interestingly , the second term proportional to @xmath83 is a multiplicative combination of the spin - orbit term and the staggered sublattice potential .
this term corresponds to the additional , electric field induced spin - orbit term that we have found in the invariant expansion model ( proportional to @xmath53 ) .
the full tight - binding hamiltonian is then given by @xmath84 the terms given above will reproduce our hamiltonian derived by symmetry analysis , eq .
( [ eq : ham2 ] ) , when being expanded around the @xmath0-points . to be able to estimate the coupling constants introduced above , we now briefly discuss and extend a tight - binding model presented in ref
. @xcite including @xmath34 and @xmath18-bands in silicene using the basis @xmath85 nearest - neighbor hopping matrix elements between given orbitals can then be expressed by parameters @xmath86 , @xmath87 , and @xmath88 as functions of slater bond parameters @xmath89 , @xmath90 , @xmath91 and curvature angle @xmath9 with @xmath92 ( see ref .
@xcite for details of the modeling ) . beyond the analysis done in ref .
@xcite , we consider the influence of an electric field @xmath39 applied perpendicularly to the atomic plane . a staggered sublattice - hamiltonian is then induced that takes in the basis ( [ eq : tbbasis1 ] ) the following form @xmath93 as only transitions between @xmath80 and @xmath45 orbitals of the same site are allowed @xcite . in this equation , @xmath94 is the charge of an electron and @xmath95 the stark element weighting the transition .
spin - orbit coupling is now included in the same way as in ref .
@xcite by the term @xmath96 where @xmath97 is the angular momentum vector , @xmath59 the spin operator , and @xmath98 the coupling constant .
the tight - binding model of @xmath34 and @xmath18 bands allows us to estimate some of the parameters introduced in the previous section as a function of the parameters @xmath99 ( that are in principle known ) as well as the stark element @xmath95 . to do so
, we now apply a unitary transformation @xmath100 to the hamiltonian that corresponds to the following change of basis @xmath101 with mixed orbitals , for example , @xmath102 .
then , the transformed hamiltonian @xmath103 can be analyzed perturbatively on the first @xmath104 block corresponding to the basis @xmath105 , which is expected to describe energy eigenstates near the fermi energy @xcite .
listing only terms including the external electric field , we find in first and second order perturbation theory the following three terms @xmath106 with parameters @xmath107 in the last step , the expressions were approximated under the assumption of low buckling , meaning @xmath108 and @xmath109 , where only the term of lowest order in @xmath88 was kept .
the term proportional to @xmath78 represents the on - site hopping occuring only in a buckled structure . with the one proportional to @xmath83 , an electric field induced term of first order in soi @xmath98
is generated that is related to a next - nearest neighbor hopping .
it depends linearly on @xmath88 , so this term is specific to the buckled silicene structure as well .
finally , the term proportional to @xmath110 is the well - known rashba spin orbit coupling reported already in refs .
all terms given above agree nicely with our previous group theoretical analysis .
we now estimate the size of the spin - orbit terms coupling to an external electric field in silicene .
the bond parameters @xmath99 , the energy @xmath111 , the angle @xmath112 , the lattice constant @xmath13 , as well as the spin orbit interaction @xmath98 are adapted from ref . @xcite .
we then approximate the stark - element @xmath113 as transition matrix elements between atomic wave functions , where in silicene we have @xmath114 , since the outer shells are provided by the @xmath115 and @xmath116 orbitals .
the corresponding wave functions were chosen as the wave functions of the hydrogen atom with the same quantum numbers .
we then find @xmath117 in silicene , where @xmath118 is the bohr radius and an outer electron experiences an effective atomic charge of @xmath119 @xcite due to screening .
hence , we estimate @xmath120 . for typical values of @xmath121
, we calculate @xmath122 thus , the additional term proportional to @xmath83 is small compared to the term proportional to @xmath78 , but similar in magnitude as the kane - mele - rashba term proportional to @xmath110 .
triggered by the theoretical prediction @xcite and experimental discovery @xcite of the quantum spin hall ( qsh ) effect , the study of topological effects in the physics of bloch bands has been a major focus of condensed matter physics in recent years @xcite .
the qsh state is a bulk insulating state featuring metallic edge states that are protected by time reversal symmetry . due to kramers
theorem , these edge states appear in so called helical pairs .
the bulk energy gap in the original proposal for the qsh effect in graphene by kane and mele @xcite is due to an intrinsic soi which preserves a residual @xmath123 spin symmetry . in the presence of this global spin quantization axis ,
the helical edge states are characterized by a perfect locking of spin and momentum : states with opposite spin move with opposite chirality along the edge . in the presence of ( rashba ) soi ,
no spin symmetry is present resulting in the absence of a global spin quantization axis .
rather , the spin quantization axis of the edge states can precess spatially .
however , as trs is present the kramers pair of helical edge states is still protected from hybridizing , i.e. , from a gap opening on the edge .
this statement is true as long as the bulk gap is maintained implying that the system is adiabatically connected to the @xmath123 preserving case and hence is still in the qsh state .
however , the role of rashba soi for the qsh effect in graphene is only detrimental : upon increasing the strength of rashba soi , the system will go through a quantum phase transition destroying the qsh phase .
conversely , given the lattice symmetry of graphene , switching on rashba spin orbit coupling can not generate the qsh phase starting from a semi - metallic or trivial insulator phase .
the slightly reduced symmetry of silicene stemming from its buckled structure entails several new couplings , at least two of which are of key relevance regarding qsh physics .
first , let us consider the staggered potential distinguishing sublattice a and b that has been introduced formally in ref .
@xcite to open a trivial gap .
while this term is symmetry forbidden in graphene , it has been shown @xcite to be induced by a simple out of plane electric field in silicene as we confirmed in our symmetry analysis above .
this provides a knob to experimentally switch off the qshe .
second , and even more interestingly , it has very recently been conjectured @xcite that the qshe can be generated by virtue of soi terms that are symmetry forbidden in graphene but allowed in silicene .
the authors of ref .
@xcite probe the sub - gap conductance as a fingerprint of the quantum spin hall state . while a non - vanishing conductance in a phase with a bulk gap is a promising signature of the qsh state , it is not in one to one correspondence with the @xmath1-invariant defining the quantum spin hall effect @xcite .
for example , what is called the qshe2-phase in ref .
@xcite is a trivial insulator which only shows the characteristic conductance of @xmath124 expected for the qsh phase since one additional pair of edge states does not contribute due to a mini - gap at the ribbon sizes considered in ref .
@xcite . in the thermodynamic limit
, the conductance in this parameter regime would be @xmath125 signaling a @xmath1-trivial phase .
the main purpose of this section is to calculate the correct phase diagram of silicene in the presence of the ( rashba ) soi terms by rigorous calculation of the @xmath1 invariant in the absence of any symmetries besides trs .
we note that our method can be applied to study the entire parameter space of the silicene band structure without any further complications .
however , in this work , we would like to focus on a particularly interesting parameter regime where the qsh phase is generated by ( rashba ) soi .
previous work on the topology of the qsh state in silicene in the presence of ( rashba ) soi @xcite has been focused on effective models valid close to the @xmath0-points .
however , in the presence of ( rashba ) soi , the bulk gap can close away from the @xmath0-points , a phase transition which would be missed by such power series expansions .
our analysis does not suffer from these limitations since we directly calculate the @xmath1-invariant characterizing the qsh phase from its very definition @xcite .
we find that there is indeed a qsh phase in the absence of the original kane - mele term @xmath61 .
this qsh effect can be switched on by only tuning silicene specific @xmath126 symmtery breaking soi terms and the staggered potential all terms that are known to have only detrimental effect as to qsh physics in graphene .
our analysis , hence , settles in the affirmative the discussion on whether other soi terms than the @xmath126 conserving kane - mele term can in principle generate a qsh state .
let us briefly give an idea of how the topology of silicene may depend on an external electric field , orienting ourselves along the lines of ref .
we consider again the effective hamiltonian of eq .
( [ eq : ham2 ] ) , derived by analytical expansion around the high - symmetry @xmath0-points . without electric fields , the energy spectrum at @xmath0 exhibits a gap of size @xmath127 .
interestingly , in the presence of soi , the bulk gap can be closed by an increasing electric field perpendicular to the plane .
the gap closes approximately at the very @xmath0-points , only if the corresponding parameter @xmath110 is small compared to the transfer energy @xmath128 , so @xmath129 .
the low - energy hamiltonian allows to give an analytical expression for the gap - closing critical field @xmath130 @xcite . at this point
we may expect a topological phase transition from a qsh to a trivial insulating phase .
this is indicated by the fact , that a gap is reestablished at the @xmath0-points , if the electric field is further increased , exceeding @xmath130 .
however , a rigorous exploration of prospective topological quantum phase transitions in the silicene parameter space requires the calculation of a topological bulk invariant .
therefore , we now turn to the general calculation of the topological @xmath1-invariant defining the qsh phase @xcite . while the @xmath1-invariant is of course a gauge invariant quantity by definition , the original literature @xcite did not provide a constructive recipe for its numerical calculation .
this is so because a calculation following the original definition requires a macroscopic gauge , though giving a gauge invariant result . by macroscopic gauge ,
we mean that the phase relation between bloch functions at remote points in the brillouin zone has to be fixed . however ,
when the band structure is calculated numerically , such phase relations are typically not accessible thus preventing the direct calculation of the invariant .
this problem has only been resolved rather recently @xcite by a more constructive recipe for the direct calculation of the @xmath1-invariant . here ,
we follow the method introduced by prodan @xcite which makes use of the elegant and manifestly gauge invariant formulation of the adiabatic connection @xcite originally introduced in a seminal work by kato back in 1950 @xcite .
recently , the manifestly gauge invariant formulation of topological invariants has been generalized to other topological band structures @xcite .
the crucial step of this approach consists in going from the bloch functions of the occupied bands to the projection operator @xmath131 onto the occupied states . as already pointed out by kato in ref .
@xcite , the advantage of this construction is that the projection operator is obviously basis ( gauge ) independent .
the adiabatic connection is defined as @xmath132,\ ] ] where @xmath133 is the derivative in momentum space .
note that this connection is manifestly independent of the basis choice within the occupied bands in contrast to the more familiar berry connection .
therefore , the adiabatic connection can be calculated numerically in a straight forward way which is in general not possible for the berry connection . in ref .
@xcite , the @xmath1 invariant of the qsh state is constructed in a similar way to ref . @xcite but using the adiabatic connection instead of the berry connection .
we refer the reader to ref .
@xcite for this both accessible and fairly self contained explicit construction and review only the resulting expression for the @xmath1-invariant @xmath134 for a rectangular lattice with lattice constants @xmath135 , for completeness , @xmath136 here , @xmath137 are the boundaries of the brillouin zone , @xmath138 is the matrix of the trs operation which is anti - symmetric at the time - reversal invariant momenta ( trim ) , and pf denotes the pfaffian .
we note that @xmath139 defines the non - trivial qsh phase whereas @xmath140 for a trivial insulator
. the parameter @xmath63 will be explained in detail below and @xmath141 describes the adiabatic evolution along the straight line from @xmath142 to @xmath143 with the path ordering operator @xmath144 .
unitaries generated by the adiabatic connection such as eq .
( [ eqn : katoprop ] ) can be conveniently calculated numerically as products of projection operators @xcite .
explicitly , @xmath145 the path ordering appearing in eq .
( [ eqn : katoprop ] ) then amounts to the ordering of the product in eq .
( [ eqn : projprod ] ) from the right to the left with increasing @xmath76 .
the construction resulting in eq .
( [ eqn : projprod ] ) is similar to a trotter decomposition for a time evolution operator and is correct to leading order in @xmath146 .
the gauge invariance of eq .
( [ eqn : prodan ] ) might not be obvious at first glance due to the basis dependence of the pfaffians . however , the combinations @xmath147 with @xmath148 appearing in eq .
( [ eqn : prodan ] ) are indeed gauge invariant up to the choice of the branch of the multivalued square - root as is obvious from the right hand side of eq .
( [ eqn : sqrtdet ] ) . that is the point where @xmath63 comes into play .
when calculating @xmath149 it has to be assured that the branch choices of the square - root at @xmath150 and @xmath151 are continuously connected to each other .
this can be done by continuously interpolating the phase factor @xmath152 between @xmath150 and @xmath151 @xcite .
if this phase has a winding such that the standard branch cut between riemann sheets ( line @xmath153 $ ] in the complex plane ) is crossed @xmath154 times during this interpolation , the naive result for @xmath134 is corrected by the factor @xmath155 . while this procedure might be numerically challenging close to critical points or for large super - cells in disordered systems , it is basically a straightforward recipe . note that the arbitrary basis choice for the representation matrix @xmath138 does not require any knowledge about the relative phase between the basis vectors at different points in @xmath156-space . furthermore
, the mentioned phase interpolation does not require any numerically inaccessible information either , since the phase factor @xmath152 at every @xmath157 is a gauge invariant quantity .
on the one hand , the procedure for the calculation of the @xmath1-invariant just described is general but requires a rectangular lattice whereas silicene crystalizes in a honeycomb lattice . on the other hand , the result for the topological invariant can not depend on the choice of the unit cell
. therefore , we will now introduce a rectangular super cell which immediately allows us to apply the above method . to this end , the common basis vectors given in eq .
( [ eq : myunitvectors ] ) are combined to a new set of basis vectors @xmath158 spanning a rectangular lattice with four atoms per site @xmath3 , @xmath4 , @xmath159 and @xmath160 , as shown in fig .
[ fig : unitc ] .
the brioullin zone is again rectangular with basis vectors @xmath161 unit cells of the two - dimensional honeycomb lattice . on the left hand side ,
the minimal unit cell is drawn , containing two basis atoms @xmath3 and @xmath4 . on the right hand side
, the unit cell has a rectangular form of doubled size . here
, four basis atoms @xmath3,@xmath4 , @xmath159 , and @xmath160 are included.,scaledwidth=60.0% ] the bloch hamiltonian of size @xmath162 associated with our tight - binding model consists of the following contributions @xmath163 the form of the @xmath156-dependent matrices is listed in eqs .
( [ eq : ah1 ] ) to ( [ eq : ah6 ] ) in [ sec : appb ] . in the extended unit cell , the @xmath0-points ( former corner points of the brillouin zone )
are mapped onto points @xmath164 inside the rectangle forming the new brillouin zone . for the hamiltonian we chose the basis @xmath165 , where the wavefunctions are of the form @xcite @xmath166 with @xmath167 . here , @xmath168 are atomic wavefunctions , in particular @xmath169 .
@xmath170 denote lattice vectors in real space , connecting the reference points of unit cells , while @xmath171 are the positions of atoms labeled @xmath172 , in the unit cell @xmath76 , relative to the reference point .
the sum runs over all cells of the crystal .
having constructed a rectangular lattice for silicene with four atoms per unit cell ( see fig . [
fig : unitc ] ) , we now investigate its topological phase diagram by direct evaluation of eq .
( [ eqn : prodan ] ) .
the unitary adiabatic time evolutions ( see eq .
( [ eqn : katoprop ] ) ) appearing in eq .
( [ eqn : prodan ] ) are evaluated numerically using eq .
( [ eqn : projprod ] ) .
the numerical calculations are done with a @xmath156-space discretization mesh of @xmath173 ( see eq .
( [ eqn : projprod ] ) ) and @xmath174 steps for the interpolation determining the phase winding @xmath63 ( see discussion below eq .
( [ eqn : sqrtdet ] ) ) .
close to the critical points , we have increased both discretizations from @xmath174 up to @xmath175 steps to reach convergence . to ensure a regular evolution of the phase factor
, we set the numerical threshold for the absolut value of @xmath176 to be greater than @xmath177 .
+ for notational simplicity , the electric field @xmath39 is absorbed in the parameters @xmath178 , that is @xmath179 .
all parameters are measured in units of the hopping matrix element @xmath128 , which is known to be of the order of @xmath180 in silicene @xcite .
we keep in mind that @xmath181 , @xmath83 , and @xmath110 depend on the external electric field .
we first benchmark the general method in a parameter regime where we have a good intuition for the expected qsh physics from previous literature @xcite .
let us start with just one non - vanishing parameter @xmath128 , all other parameters being zero .
this choice corresponds to the well known gapless dirac cones at the @xmath0-points . upon switching on @xmath61 @xcite ,
the system exhibits a bulk gap of size @xmath182 and we reproduce @xmath139 @xcite .
we now add a term proportional to @xmath181 depending on the external field . upon increasing @xmath181
the bulk gap closes at the @xmath0-points for a critical value as predicted before . for even stronger fields a trivial gap opens ,
i.e. , @xmath183 . similarly
, the gap closes for terms proportional to @xmath83 .
next , we add the rashba - term proportional to @xmath110 and the intrinsic soi - term proportional to @xmath62 . the first term @xmath110 tends to establish three additional minima of the energy gap , that are shifted away from the @xmath0-point .
on the other hand , @xmath62 , primarily providing gapless states at the corners of the brillouin zone ( bz ) , causes a modification of the band slope . from an analysis of the overall band structure , one finds , that the interplay of both terms @xmath110 and @xmath62 may possibly close the band gap away from the @xmath0-points ( see fig .
[ fig : threemingap ] ) .
+ plot of the bulk band structure of silicene with nonzero parameters @xmath184 , @xmath185 and @xmath186 .
a rectangular unit cell is used , as explained in the text . here , only half the brioullin - zone
is plotted due to implications of trs . under this parameter choice
, the bulk band gap is about to close away from the @xmath0-points , which correspond to the points @xmath187 ( red dot).,title="fig : " ] + after these general considerations , we would now like to focus on the interesting parameter regime identified in ref .
@xcite , i.e. , we consider @xmath188 as free parameters that are measured in units of @xmath128 .
all other parameters are set to zero . unfortunately , the analysis in ref .
@xcite was not suitable to predict the correct phase diagram for the @xmath1 invariant .
however , our more rigorous analysis confirms the general conjecture that in this regime , a combination of @xmath110 and @xmath62 is able to drive a topological quantum phase transition away from the @xmath0-points which induces a non - trivial qsh phase .
if we only switch on @xmath181 , a trivial gap of size @xmath189 opens . by tuning @xmath110 and @xmath62
, the bulk gap can be closed away from the @xmath0-points , and a gap characterized by @xmath139 , i.e. , a qsh state emerges .
we present the phase diagram in the identical parameter regime as presented in ref .
@xcite in fig .
[ fig : phases1 ] . our direct calculation of the @xmath1-invariant disagrees with ref .
@xcite both qualitatively ( absence of the qshe2 ) and quantitatively ( significantly different phase boundary for the qsh phase ) . however , we would again like to point out that we can definitely confirm the phenomenology of a qsh phase in the absence of the kane - mele term @xmath61 . instead , the extended qsh region shown in fig .
[ fig : phases1 ] is only driven by the two silicene specific ( rashba ) soi terms @xmath110 and @xmath62 which is conceptually very interesting .
the existence of gapless edge states was additionally verified by the simulation of zigzag - edge silicene nanoribbons , as shown in fig .
[ fig : qshe12 ]
. however , we emphasize , that our topological analysis is based on bulk properties and does not depend on specific forms of the boundaries .
+ lastly , we found the phase diagram in fig .
[ fig : phases1 ] to be robust against small perturbations proportional to the term @xmath61 . for larger kane - mele parameters @xmath190 ,
another non - trivial regime @xmath191 enters the phase diagram .
this regime occurs independently of @xmath62 and can be expelled again by increasing @xmath110 .
essentially , the previously described quantum phase transitions are not affected .
+ topological phase diagram of silicene with free parameters @xmath192 .
the staggered potential @xmath193 is fixed as well as @xmath194 .
all couplings are measured in units of the hopping energy @xmath128 .
the white points close to the phase boundary denote critical regions where the bulk gap is so small that our numerical calculation did not converge properly.,title="fig : " ] + simulation of the band dispersion of a silicene nanoribbon with zigzag - edges for two selected points of different regimes in parameter space .
the unit cell corresponding to the width of the ribbon contains @xmath195 atoms .
the numerical calculation was based on a discretization of the @xmath156-space in @xmath196 steps .
( a ) @xmath184 , @xmath197 and @xmath198 .
( b ) @xmath184 , @xmath199 and @xmath200 . in the qsh - regime ( a )
, there is an odd number of pairs of edge states at each edge . on the other hand ,
an additional pair of edge states leading to an even number of pairs , causes the topology to be trivial in the regime represented by ( b).,title="fig:",scaledwidth=110.0% ] + ( a ) simulation of the band dispersion of a silicene nanoribbon with zigzag - edges for two selected points of different regimes in parameter space .
the unit cell corresponding to the width of the ribbon contains @xmath195 atoms .
the numerical calculation was based on a discretization of the @xmath156-space in @xmath196 steps .
( a ) @xmath184 , @xmath197 and @xmath198 .
( b ) @xmath184 , @xmath199 and @xmath200 . in the qsh - regime ( a )
, there is an odd number of pairs of edge states at each edge . on the other hand
, an additional pair of edge states leading to an even number of pairs , causes the topology to be trivial in the regime represented by ( b).,title="fig:",scaledwidth=110.0% ] + ( b )
a symmetry analysis of the lattice of silicene close to the @xmath0 points of the brillouin zone was performed . with the help of the invariant expansion model ,
the @xmath34-band hamiltonian was constructed by symmetry considerations only , including spin - orbit coupling and external electric fields perpendicular to the atomic plane .
thereby , we discovered an additional term that is allowed by symmetry and related to an interplay of spin - orbit coupling and an external electric field in a buckled honeycomb structure .
we supplemented the symmetry analysis by a tight - binding model which allowed us to estimate the relevant parameters of the model .
subsequently , we carefully analyzed the topological properties of the band structure and proved that a topological phase transition can be generated by an external electric field .
this analysis enabled us to plot a topological phase diagram of silicene employing prodan s manifestly gauge - invariant method for the direct calculation of the @xmath1 topological invariant defining the quantum spin hall phase .
since this method requires a rectangular lattice , we considered a rectangular silicene super - cell that contains four atoms .
interestingly , the tunable phase transition can happen in a destructive as well as a constructive way , i.e. , a quantum spin hall phase can not only be destroyed but also be generated by means of external ( rashba ) spin orbit coupling .
we have benefitted from discussions with roland winkler .
this work has been financially supported by the deutsche froschungsgemeinschaft ( dfg - jst research unit `` topotronics '' , priority program `` topological insulators '' ) , the european science foundation , the helmholtz foundation ( viti ) , and the swedish research council .
let us , for completeness , shortly review the general theory of the invariant expansion . in this approach ,
the hamiltonian is constructed to be invariant under all symmetry operations of the point group of the underlying lattice .
when spin is included , it is composed block - wise according to irreducible representations ( ir ) of the corresponding double group .
each block @xmath201 , that may depend on some general tensor components @xmath202 , has to reflect the symmetry properties of all the irs @xmath203 , that are contained in the product @xmath204 where the integer @xmath205 is the multiplicity .
the most general ansatz is then @xmath206 here , @xmath207 runs over all ir @xmath203 included in the product and @xmath208 are material - specific constants .
the symmetrized tensor operator components @xmath209 are parameters like the wave vector @xmath38 , as well as the external magnetic @xmath210 or electric field @xmath211 .
furthermore , @xmath212 are the symmetrized matrices we wish to find to construct the hamiltonian .
evidently , @xmath213 und @xmath214 are indices numbering the different possible matrix- and tensor - components .
these indices run from unity to limits that depend on the order of the expansion .
one set of the @xmath207 , @xmath213 , and @xmath214 is called an invariant .
the irreducible tensor components belonging to an ir are composed analogously to its eigenfunctions . with the help of projection operators
@xcite , that contain the matrix representations of each ir , we can combine tensor components at any order and project them onto the appropriate ir of the point group . to construct the basis matrices @xmath212 ,
we can use the wigner - eckart theorem @xcite . since any point group is a subgroup of the full rotation group @xmath215 , the eigenfunctions of any ir of the point group are also eigenfunctions of @xmath215 and
can be written in terms of spherical harmonics .
angular momentum quantum numbers are assigned to each ir , by defining the axial vector components @xmath216 , @xmath217 , and @xmath218 to be the real angular momentum eigenstates with quantum number @xmath219 .
each ir can now be classified with angular momentum quantum numbers by comparing its eigenfunctions to a table of real spherical harmonics .
a symmetrized matrix @xmath220 ( that transforms like the ir @xmath221 of dimension @xmath222 and numbering @xmath223 ) which is contained in the block @xmath224 is then derived by clebsch - gordan coefficients ( cgc ) @xmath225 where the order of @xmath21 , @xmath76 , and @xmath156 is crucial . in eq .
( [ eq : ccbasmatr ] ) , @xmath226 is again the multiplicity , that shows how often the ir @xmath227 is included in the product @xmath224 . for multiplicities larger than one ,
we find linear independent sets of basis matrices .
we now start to perform an invariant expansion of the group @xmath32 , that we found to be the group of the wave vector at the @xmath0-points in silicene ( see ref .
@xcite for a general discussion of this point group ) .
we are interested in the two - dimensional @xmath34-bands of silicene .
the orbital part of the wavefunction corresponds to the two - dimensional ir @xmath35 of the group @xmath32 .
since we wish to include spin in our symmetry analysis , we have to complement this ir by @xmath36 , the appropriate double group ir of @xmath32 .
consequently , we start from the following product of representations @xmath228 the resulting @xmath104 hamiltonian based on the invariant expansion can then be composed in blocks of four @xmath229 matrices @xmath230 the construction by cgcs ( as discussed in the previous section ) implies that this @xmath34-band hamiltonian is given in the basis @xmath231 , with @xmath232 and @xmath42 , @xmath43 .
each block of the hamiltonian is thus decomposed into irs .
general tensor components of any desired order can be found with the help of projection operators from the point group character table . in table
[ tab : expirrcompd3 ] , we list components of interest up to third order . @ llllp2.4cmp4 cm & @xmath233 & @xmath234 & @xmath235 & & + @xmath236 & @xmath237 & @xmath237 & @xmath237 & & @xmath238 ; @xmath239 + @xmath240 & @xmath237 & @xmath237 & @xmath241 & @xmath218 ; @xmath8 & + @xmath35 & @xmath242 & @xmath241 & @xmath243 & @xmath244 ; @xmath245 & @xmath246;@xmath247 + + @ llllp6.85 cm & @xmath233 & @xmath234 & @xmath235 & + @xmath236 & & & & @xmath248 ;
@xmath249 + @xmath240 & & & & @xmath250 + @xmath35 & & & & @xmath251 ; @xmath252 + identifying combinations of general tensor components of @xmath253 with angular momentum quantum numbers , we can assign such quantum numbers to the irs itself .
the basis matrices are then identified from cgcs according to eq .
( [ eq : ccbasmatr ] ) and listed in table [ tab : expbasmatrd3 ] .
note that we have applied a unitary basis transformation to guarantee a more symmetric basis @xmath254 .
this transformation is done for comparison with the graphene case carefully analyzed in ref .
@xcite . @ ll p4 cm @xmath255 & @xmath256 & @xmath257 @xmath258
@xmath259 + @xmath260 & @xmath261 & @xmath262 @xmath263 + @xmath264 & @xmath265 & @xmath266 + we use the symmetrized matrices and tensor components listed in tables [ tab : expirrcompd3 ] and [ tab : expbasmatrd3 ] to expand the hamiltonian up to first orders in @xmath38 and the electric field @xmath39 perpendicular to the plane . note that since @xmath267 is hermitian , the coefficients of the diagonal blocks have to be real , while those of the off - diagonal blocks can in principle be imaginary .
we write @xmath268 and @xmath269 to include both cases with real coefficients @xmath268 , @xmath270 .
then , we obtain for the hamiltonian @xmath271 @xmath272 @xmath273 for a better physical interpretation , we once more change the basis by an unitary transformation and present the total hamiltonian in the basis @xmath41 . after this basis transformation , the pauli matrices @xmath18 and
@xmath45 obtain the following physical meaning : @xmath18 acts on the space of sublattices @xmath3 and @xmath4 , and @xmath45 on the spin space . in lowest order
, the hamiltonian in our new basis exhibits sixteen terms labeled by coefficients @xmath274 to @xmath54 @xmath275 + \nonumber \\ & a_{10 } e_z [ \sigma_x(s_x k_y + s_y k_x)+\sigma_y(s_x k _ x -s_y k_y ) ] + \nonumber \\ & a_{11 } e_z \sigma_0 ( s_x k_y - s_y k_x ) + a_{12 } e_z ( \sigma_x k_y - \sigma_y k_x ) s_0 + \nonumber \\ & a_{13}\sigma_z ( s_x k_y - s_y k_x ) + a_{14 } ( \sigma_x k _ y - \sigma_y k_x ) s_z + \nonumber \\ & a_{15 } e_z \sigma_z ( s_x k_x+s_y k_y)+a_{16}e_z(\sigma _ x k_x + \sigma_y k_y)s_z . \nonumber\end{aligned}\ ] ] note , that the coefficients in eqs .
( [ eq : h44grc3v ] ) to eq .
( [ eq : h45grc3v ] ) and in eq .
( [ eq : ham ] ) are consistent , in the sense , that they are related by mutual linear combinations . here
, we would already like to point out the additional , interesting spin - orbit term proportional to the coefficient @xmath53 , coupling directly the out - of - plane components of spin and electric field .
the terms derived so far in the invariant expansion have not yet been checked for consistency with time reversal symmetry which holds throughout this work since we only consider the influence of electric fields .
importantly , the point group @xmath276 provides symmetry operations that map the inequivalent corner points @xmath0 and @xmath19 of the brillouin zone onto each other .
for example , we can consider the reflection at a plane perpendicular to the atomic plane and including the y - axis ( called @xmath217 in ref .
@xcite ) . this symmetry operation has the following impact on the hamiltonian @xmath277 where @xmath278 is the matrix representation of the operation @xmath279 .
polar ( @xmath280 ) and axial ( @xmath253 ) tensor components will then transform like @xmath281 and @xmath282 .
the sublattices will not be changed by @xmath217 . in a compact notation
, we can write @xmath283 . with the help of @xmath217 we find the form of the hamiltonian at one of the two inequivalent @xmath0 points .
all terms , that change sign under @xmath217 will thus be modified by the additional parameter @xmath284 to mark the difference between the valleys @xmath0 and @xmath19 .
the true time reversal operator equals @xmath285 , where @xmath286 is the complex conjugation operator @xcite and @xmath287 a pauli matrix corresponding to the valley isospin .
its impact on the hamiltonian can be written as @xmath288 where @xmath289 depending on whether the tensor component @xmath202 changes sign under time reversal or not . hence , both operators @xmath217 and @xmath290 lead to transformations between the valleys .
we now combine both of them to a ( new ) time - reversal operator within a single valley @xmath291 ( in the spirit of ref .
@xcite ) : @xmath292 .
this operator yields an additional symmetry constraint for all terms @xmath293 which forces the coefficients @xmath294 , @xmath295 , @xmath296 , and @xmath297 in eq .
( [ eq : ham ] ) to vanish .
our result is the low - energy hamiltonian of silicene near the @xmath0-points ( in first orders in @xmath38 and @xmath39 ) presented in the basis @xmath41 .
it is given by @xmath298 \nonumber+\\ & a_{11 } e_z \sigma_0 ( s_x k_y - s_y k_x ) + a_{13}\sigma_z ( s_x k_y - s_y k_x ) + \nonumber \\ & a_{16}e_z(\sigma_x k_x + \tau_z\sigma_y k_y)s_z .
\nonumber\end{aligned}\ ] ]
the terms contributing to the tight - binding hamiltonian in eq .
( [ eq : htotal ] ) , in the basis @xmath165 , take the explicit form @xmath299 @xmath300 @xmath301 @xmath302 @xmath303 @xmath304 the matrix elements are given in tables [ tab : bighamco ] and [ tab : bighamco2 ] . + @ llll @xmath305 & @xmath306 & @xmath307 & @xmath308 + @xmath309 & @xmath310 & @xmath311 & @xmath312 + @xmath313 & @xmath314 & @xmath315 & @xmath316 + @xmath317 & @xmath318 & & + @ l p6.9 cm
l p6.9 cm @xmath319 & @xmath320 & @xmath321 & @xmath128 + @xmath322 & @xmath323 $ ] & @xmath324 & @xmath325 + @xmath326 & @xmath327 & @xmath328 & @xmath329 $ ] + @xmath330 & @xmath331 & @xmath332 & @xmath333 + @xmath334 & @xmath335 & @xmath336 & @xmath337 + @xmath338 & @xmath339 & @xmath340 & @xmath341 + + note , that coefficients are related to parameters of the group theoretical approach by @xmath342
10 dresselhaus m s , dresselhaus g and jorio a 2008 ( heidelberg new york : springer verlag ) sakurai j j 1987 ( addison wesley ) koster g f , dimmock j o , wheeler r g and statz h 1963 ( cambridge , ma : mit ) beenakker c w j 2008 pp 13371354 | silicene consists of a monolayer of silicon atoms in a buckled honeycomb structure .
it was recently discovered that the symmetry of such a system allows for interesting rashba spin - orbit effects .
a perpendicular electric field is able to couple to the sublattice pseudospin , making it possible to electrically tune and close the band gap .
therefore , external electric fields may generate a topological phase transition from a topological insulator to a normal insulator ( or semimetal ) and vice versa .
the contribution of the present article to the study of silicene is twofold : first , we perform a group theoretical analysis to systematically construct the hamiltonian in the vicinity of the @xmath0 points of the brillouin zone and find an additional , electric field induced spin - orbit term , that is allowed by symmetry .
subsequently , we identify a tight binding model that corresponds to the group theoretically derived hamiltonian near the @xmath0 points .
second , we start from this tight binding model to analyze the topological phase diagram of silicene by an explicit calculation of the @xmath1 topological invariant of the band structure . to this end , we calculate the @xmath1 topological invariant of the honeycomb lattice in a manifestly gauge invariant way which allows us to include @xmath2 symmetry breaking terms like rashba spin orbit interaction into the topological analysis .
interestingly , we find that the interplay of a rashba and an intrinsic spin - orbit term can generate a non - trivial quantum spin hall phase in silicene .
this is in sharp contrast to the more extensively studied honeycomb system graphene where rashba spin orbit interaction is known to compete with the quantum spin hall effect in a detrimental way . |
relative intensity squeezing by four - wave mixing in an atomic vapor is emerging as a promising technique for performing high - precision measurements beyond the shot - noise limit .
first demonstrated by mccormick et al .
@xcite , the technique uses atomic coherences to produce quantum correlated `` twin beams '' , enabling the shot - noise of one beam to be measured and subtracted from the other to obtain a low - noise differential measurement ; for example of a weakly absorbing sample .
this scheme was recently shown to reduce the relative intensity noise by @xmath2db below the shot - noise limit @xcite , and noise reduction has been observed in both the low fourier frequency @xcite and multi - mode imaging @xcite domains .
furthermore , as one of the twin beams is near - resonant with the atoms , this squeezing technique has promising applications in quantum information processing @xcite . however , absorption near resonance degrades the quantum correlations . both mixing gain and absorption losses
occur simultaneously as the beams propagate through the vapor , and are therefore competing processes .
earlier theoretical investigations of this system have applied numerical methods @xcite and the heisenberg - langevin formalism @xcite to predict the resulting degree of squeezing .
the numerical model demonstrated excellent agreement with experimental results , but it can be difficult to gain insight into the competing processes from numerical calculations .
the heisenberg - langevin model provided a microscopic description of a specific four - wave mixing configuration in a cold atomic gas , which accurately predicted the resulting gain profiles .
however , calculation of the predicted squeezing required complex matrix integrals and no comparison to experimentally measured squeezing was presented . in this work
, we present a very general approach for determining the squeezing produced by a four - wave mixing system , and develop a matrix - based analysis method to include arbitrarily many injected vacuum modes .
considering special cases , simple closed - form expressions are easily obtained .
finally , we present experimentally measured squeezing from four - wave mixing in a rubidium-85 vapor , and demonstrate how the model can be used as a diagnostic tool to determine the limiting technical factors .
the `` double-@xmath3 '' four - wave mixing scheme introduced by mccormick et al @xcite uses a high - intensity `` pump '' beam to drive a cycle of four off - resonant transitions in a hot rubidium vapor , causing the emission of correlated `` probe '' and `` conjugate '' photons ( fig .
[ fig : fwm]a ) .
the probe transition is stimulated by a seed laser incident at an angle @xmath4 to the pump , resulting in the spontaneous emission of the conjugate on the opposite side of the pump beam ( fig .
[ fig : fwm]b ) .
the beam powers are measured individually and subtracted to obtain the relative intensity noise as measured on a spectrum analyzer ( s.a . ) .
labelling the fock - space annihilation operators of the probe , conjugate and pump by @xmath5 , @xmath6 and @xmath7 respectively and the interaction strength by @xmath8 , the interaction picture hamiltonian is @xmath9 in the `` undepleted pump '' approximation , the intense pump beam remains in its initial coherent state @xmath10 and the substitution @xmath11 can be made : @xmath12 the time - evolution of this hamiltonian over the interaction time - scale @xmath13 is @xmath14 this is the two - mode squeezing operator for modes @xmath5 and @xmath6 , where @xmath15 is the `` squeezing parameter '' @xcite .
the four - wave mixing system therefore produces a two - mode squeezed state , reducing amplitude difference noise at the expense of increasing phase difference noise @xcite . the phase of @xmath15 results in a rotation of the ( arbitrary ) measurement quadratures , so @xmath15 may be taken as real and positive .
the probe and conjugate modes @xmath5 and @xmath6 are then transformed as @xmath16 defining the number operator of the incident probe beam as @xmath17 and making the bright beam approximation @xmath18 , the number operators after squeezing become @xmath19 where @xmath20 is the increase in probe intensity , termed the `` mixing gain '' .
the relative intensity operator @xmath21 is unchanged by @xmath22 , so .
hence the beams have been amplified without increasing the relative intensity noise ; they are relative intensity squeezed .
the noise figure of the process ( or `` degree of squeezing '' ) is the ratio of the measured noise to the corresponding shot - noise level for equal optical power .
assuming an initially shot - noise limited probe , the noise figure is @xmath23 therefore the measured noise power can be reduced arbitrarily below the shot - noise limit in the limit of ideal detection .
however , optical losses are unavoidable and occur both within the medium ( e.g. absorption ) and after it ( e.g. imperfect detection ) .
these losses randomly eject photons from the probe and conjugate beams , decorrelating their intensities and degrading the observed degree of squeezing .
we now construct models to quantify this effect .
we initially consider only losses that occur after mixing , such as from imperfect optical transmission or detection efficiency .
these losses are modelled by a beamsplitter with an empty port @xcite whose output state is a combination of the input and vacuum modes , contributing `` vacuum fluctuations '' to the transmitted beam @xcite . denoting the vacuum modes introduced by losses on the probe and conjugate by the annihilation operators @xmath24 and @xmath25 respectively , the standard beam - splitter input - output relations @xcite give @xmath26 where @xmath27 and @xmath28 are the fractions of the probe and conjugate intensities transmitted .
the relative intensity noise can then be expressed in terms of the individual beam variances and covariance to give @xmath29 computing the variances using eq .
, the noise figure corresponding to four - wave mixing followed by optical losses is @xmath30 this expression highlights the importance of balanced beam detection , as unbalanced losses ( @xmath31 ) result in detection of amplified noise instead of squeezing .
the four - wave mixing process consists of raman transitions between the hyperfine ground states ( fig . [
fig : fwm]a ) , which are most efficient when the intermediate virtual level is tuned close to resonance .
however , this also increases direct absorption from the doppler broadened transition , increasing losses and reducing correlations . to analyze this trade - off ,
we develop a model for the effect of competing mixing and absorption on relative intensity squeezing . following the approach of the numerical model presented in ref .
@xcite , the beam trajectories through the medium are divided into @xmath32 discrete interleaved stages of gain and loss ( fig .
[ fig : glmodel ] ) . distributed models of this type were first proposed by loudon @xcite , and applied by caves and crouch @xcite to model distributed squeezing losses in a single - mode parametric amplifier .
we present a fully analytical model of the relative squeezing produced by this system in the continuum limit , including losses on both beams .
each stage comprises ideal squeezing ( by parameter @xmath15 ) followed by loss ( represented by transmission coefficients @xmath33 on the probe and @xmath34 on the conjugate ) .
consecutive stages of the model are related by combining eq . with eq .
to give @xmath35 the overall model is parameterized in terms of the overall squeezing parameter @xmath36 in the absence of losses , and transmissions @xmath37 and @xmath38 in the absence of squeezing .
these are related to the incremental coefficients above by @xmath39 , @xmath40 and @xmath41 .
this transformation can be written in matrix form and applied recursively to express the output beam operators @xmath42 and @xmath43 as a sum of the incident beam operators ( @xmath44 and @xmath45 ) and the injected vacuum operators ( @xmath46 and @xmath47 ) as @xmath48 { \begin{pmatrix } { \ensuremath \hat{a}}_n \\ { { \ensuremath \hat{b}}^\dagger}_n \end{pmatrix } } & = { \ensuremath \mathbf{a } } { \begin{pmatrix } { \ensuremath \hat{a}}_{n-1 } \\ { { \ensuremath \hat{b}}^\dagger}_{n-1 } \end{pmatrix } } + { \begin{pmatrix } \sqrt{1-t_a^2 } \;{\ensuremath \hat{x}}_n \\ \sqrt{1-t_b^2}\;{{\ensuremath \hat{y}}^\dagger}_n \end{pmatrix } } \\ & = { \ensuremath \mathbf{a}}^n { \begin{pmatrix } { \ensuremath \hat{a}}_0 \\ { { \ensuremath \hat{b}}^\dagger}_0 \end{pmatrix } } + \sum_{i=1}^n { \ensuremath \mathbf{a}}^{n - i } { \begin{pmatrix } \sqrt{1-t_a^2}\ ; { \ensuremath \hat{x}}_i \\ \sqrt{1-t_b^2}\ ; { { \ensuremath \hat{y}}^\dagger}_i \end{pmatrix } } , \end{aligned}\label{eq : glmatrix}\ ] ] where @xmath49 for algebraic simplicity , we introduce a polymorphic operator @xmath50 consisting of the probe annihilation and conjugate creation operators @xmath51 and @xmath52 , followed by the injected vacuum operators @xmath53 and @xmath54 for @xmath55 . expanding eq . in terms of this operator and a set of coefficients @xmath56 and @xmath57
gives @xmath58 using standard statistical identities , the relative intensity operator and its variance are @xmath59 in the bright probe beam approximation ( @xmath60 ) , this variance simplifies to @xmath61 the shot - noise limit is @xmath62 , so the degree of squeezing for an initially shot - noise limited probe with @xmath63 is @xmath64 it remains to express the @xmath56 , @xmath57 coefficients in terms of the model parameters @xmath37 , @xmath38 and @xmath36 , and hence obtain an ab - initio expression for the degree of squeezing .
equating the coefficients of eq .
and eq . leads to @xmath65 hence each of the @xmath32 vacuum modes @xmath46 contribute a term to the variance in eq . : @xmath48 ( \alpha_1 \alpha_{2i+1 } - \beta_1 \beta_{2i+1})^2 & = { \begin{pmatrix } \alpha_1 , & -\beta_1 \end{pmatrix}}{\begin{pmatrix } \alpha_{2i+1 } \\ \beta_{2i+1 } \end{pmatrix}}{\begin{pmatrix } \alpha_{2i+1 } , & \beta_{2i+1 } \end{pmatrix}}{\begin{pmatrix } \alpha_1 \\ -\beta_1 \end{pmatrix } } \\ & = { \begin{pmatrix } \alpha_1 , & -\beta_1 \end{pmatrix}}{\ensuremath \mathbf{a}}^{n - i } { \begin{pmatrix } \sqrt{1-t_a^2 } \\ 0 \end{pmatrix } } { \begin{pmatrix } \sqrt{1-t_a^2 } , & 0 \end{pmatrix}}\left({\ensuremath \mathbf{a}}^t\right)^{n - i}{\begin{pmatrix } \alpha_1 \\ -\beta_1 \end{pmatrix } } \\ & = { \begin{pmatrix } \alpha_1 , & -\beta_1 \end{pmatrix}}{\ensuremath \mathbf{a}}^{n - i } \begin{pmatrix } 1-t_a^2 & 0 \\ 0 & 0 \end{pmatrix } { \ensuremath \mathbf{a}}^{n - i}{\begin{pmatrix } \alpha_1 \\
-\beta_1 \end{pmatrix } } .
\end{aligned}\label{eq : vaccontrib}\ ] ] each operator @xmath47 contributes a term analogous to eq . , but whose diagonal matrix is . temporarily neglecting the @xmath66 contributions to the variance in eq . , and summing over the vacuum contributions gives @xmath67 the continuum behaviour is recovered in the limit @xmath68 . to obtain a closed form expression for the sum
, the infinitesimal parameters are expanded as a power series in @xmath69 .
expanding the elements of @xmath70 in eq .
gives @xmath71 similarly taking @xmath72 , the sum in braces in eq .
is @xmath73 it can be easily verified that this sum obeys the geometric series relation @xmath74 expanding to order @xmath69 using eq .
gives @xmath75 taking the limit @xmath68 , the neglected @xmath76 terms vanish and @xmath77 .
hence the sum @xmath78 converges and obeys @xmath79 this is a system of four linear equations for the elements of @xmath78 in terms of the model parameters @xmath37 , @xmath38 and @xmath36 , and can be solved algebraically .
the sum @xmath78 contains all terms in the variance of eq . except @xmath66 which correspond to the probe and conjugate coefficients .
the probe contribution is @xmath80 while the conjugate contribution @xmath81 has diagonal .
computing the full variance sum in eq .
yields @xmath82 we introduce one final stage of loss to model optical losses after mixing ( as in 4 ) , scaling each coefficient by the relevant transmission factor ( @xmath83 or @xmath84 ) and introducing the extra vacuum contributions @xmath85 .
the net variance in eq .
is therefore @xmath86 the measured beam powers relative to the incident probe power ( the `` effective gains '' ) are @xmath87 and the relevant shot - noise limit is @xmath88 .
evaluating all the contributions to eq . , we obtain an analytic expression for the degree of relative intensity squeezing produced by this system .
this expression is algebraic in @xmath37 , @xmath38 and @xmath36 , but runs to a dozen typeset lines .
however , special cases are readily derived and provide physical insight not readily accessible from numerical models .
in the experimentally studied case ( refs . [ 17 ] ) , detection efficiencies are carefully balanced ( @xmath89 ) and the far - detuned conjugate experiences negligible absorption ( @xmath90 ) .
the corresponding degree of squeezing is @xmath91 with parameters @xmath92 and @xmath93 .
the three terms describe the shot - noise limit , correlations from four - wave mixing , and injected vacuum noise . .
] figure [ fig : compareloss ] shows the noise figure as a function of the probe transmission @xmath37 and intrinsic mixing gain @xmath94 ( in excellent agreement with the numerical model of ref .
note the counter - intuitive result that the strongest squeezing is obtained with imperfect transmission ( @xmath95 ) .
this is because the shot noise carried by the incident probe beam is also amplified by the mixing process , and a small `` optimal '' level of probe loss decreases this contribution to the measured noise power before injected vacuum noise dominates .
this optimal level is easily obtained by minimizing eq . .
the model can also be applied to other four - wave mixing systems .
for example , interchanging the probe and conjugate wavelengths produces the `` reverse '' configuration ( fig .
[ fig : reverse]a ) .
this system is interesting as the weakly - coupled conjugate transition is brought closer to resonance , producing much higher intrinsic mixing gain for the same beam powers . in this case , the probe experiences negligible absorption ( @xmath96 ) compared to the conjugate ( @xmath97 ) , and the predicted squeezing is @xmath98 the vacuum noise term in eq .
is considerably larger than in eq . , resulting in several decibels difference for moderate levels of absorption ( fig .
[ fig : reverse]b ) . unlike the probe losses discussed above , losses on the conjugate only destroy correlations and introduce noise , so squeezing by four - wave mixing in the reverse configuration
is always less effective for the same level of intrinsic mixing gain . .
] finally , it is worth noting for consistency that in the limit of both @xmath37 , @xmath99 , the post - mixing optical - loss result of eq . is obtained .
the analytic model derived above provides a simple yet powerful tool for optimizing the degree of squeezing obtained experimentally . it
not only predicts which parameters will provide optimal results , but can also be used as a diagnostic tool to determine which factors are limiting the experimentally measured degree of squeezing . to demonstrate this
, we constructed a four - wave mixing apparatus as described in ref .
a @xmath100 pump beam intersects a @xmath101 probe beam with @xmath102 beam waists of @xmath103 m and @xmath104 m respectively at an angle of @xmath105 within a pure @xmath0rb vapor cell of internal length @xmath106 heated to @xmath107c ( fig .
[ fig : fwm]b ) .
the probe was generated by an aom with fixed detuning @xmath108mhz below the pump , which was scanned across the doppler broadened @xmath109 resonance at @xmath110 nm .
the relative intensity between probe and conjugate was measured with a balanced photodetector ( thorlabs pdb150a ) , refitted with high - efficiency photodiodes ( hamamatsu s3883 , net efficiency @xmath111 ) .
the overall detection efficiency of the system was @xmath112 .
the relative intensity noise was measured with a rhode & schwarz fsp7 spectrum analyzer at an analysis frequency of @xmath113mhz with @xmath114khz resolution bandwidth .
the effective probe and conjugate gains ( @xmath115 and @xmath116 ) were measured as a function of pump beam detuning ( fig .
[ fig : exper]a ) and used to simultaneously solve for the intrinsic gain @xmath117 and probe transmission @xmath37 ( fig . [ fig : exper]b ) via the coefficients @xmath118 and @xmath119 .
note that the gain resonance extends well into the doppler - broadened absorption resonance for detunings below @xmath120mhz from the line - centre , demonstrating strong competition between the processes .
rb @xmath121 transition . ]
the model predicts significantly stronger squeezing should be possible for these parameters than measured experimentally ( fig .
[ fig : exper]c ) . as our incident probe beam
was measured to be shot - noise limited , this demonstrates that our noise measurement was limited by technical factors and not by insufficient intrinsic gain or excessive absorption .
one cause for our discrepancy is the technical difficulty in eliminating the bright pump beam after the vapor cell .
detection of the pump introduces uncorrelated fluctuations and increases the measured noise level @xcite .
this is evident around zero detuning , where the probe is fully absorbed ( fig .
[ fig : exper]a ) but the measured noise is far above the standard quantum limit ( fig . [
fig : exper]d ) . upon blocking the pump beam ,
the relative intensity noise was measured at the shot - noise limit .
subtracting this background level implies that @xmath122db of squeezing would be obtained by eliminating cross - beam detection .
the model assumes that gain occurs uniformly throughout the vapor cell , which requires that the beams be overlapped over the entire region .
if the beams are not properly overlapped , the mixing strength decreases and gain becomes spatially varying . as losses are unchanged , the relative intensity noise increases as a result .
this is likely the cause of the remaining discrepancy , and improved squeezing could be achieved by manipulating beam alignment and waists to ensure proper overlap throughout the cell .
we have presented a method for analytically calculating the degree of squeezing produced by a four - wave mixing system in the presence of absorption . our model included the contributions from arbitrarily many injected vacuum modes that were subsequently squeezed by the system , producing an ab - initio quantum mechanical description of the introduced losses .
our general result is not reliant on implementation details and can be applied to analyze any four - wave mixing scheme , while the matrix methods techniques we developed can be applied to model other systems both within quantum optics and more generally .
we considered two special cases , corresponding to the experimentally studied system ( introduced in ref .
@xcite ) and the `` reverse '' configuration , with probe and conjugate wavelengths interchanged .
we presented closed - form expressions for the relative intensity noise in these cases , and demonstrated that a small level of probe loss was desirable to suppress amplification of the initial shot - noise .
the reverse configuration was shown to produce the same squeezing at equal gain for ideal transmission , but was significantly more sensitive to losses .
the model was applied to analyze experimentally measured squeezing and determine the intrinsic mixing gain and transmission factors of the four - wave mixing resonance . comparing the expected squeezing to measured results provided insight into the factors limiting our measurement and hence where to direct effort in optimizing the many free parameters of the system .
while it should be noted that this model considers the propagation of a gaussian beam mode only , an arbitrary beam can be analyzed as a product state of orthogonal spatial modes , with each mode independently squeezed @xcite .
the model established in this paper may therefore be applied to each pair of spatial modes and the resulting noise powers summed to obtain the overall relative intensity squeezing for a multi - mode beam .
such multi - mode squeezing has been experimentally demonstrated @xcite , with promising applications in quantum imaging . | four - wave mixing near resonance in an atomic vapor can produce relative intensity squeezed light suitable for precision measurements beyond the shot - noise limit .
we develop an analytic distributed gain / loss model to describe the competition of mixing and absorption through the non - linear medium . using a novel matrix calculus , we present closed - form expressions for the degree of relative intensity squeezing produced by this system .
we use these theoretical results to analyze experimentally measured squeezing from a @xmath0rb vapor and demonstrate the analytic model s utility as an experimental diagnostic . 10 c. f. mccormick , v. boyer , e. arimondo , and p. d. lett , `` strong relative intensity squeezing by four - wave mixing in rubidium vapor , '' * 32 * , 178180 ( 2007 ) .
q. glorieux , l. guidoni , s. guibal , j .-
p .
likforman , and t. coudreau , `` strong quantum correlations in four wave mixing in @xmath0rb vapor , '' ( spie , 2010 ) , vol .
7727 of _ proc .
spie _ , p. 772703 . c. f. mccormick , a. m. marino , v. boyer , and p. d. lett , `` strong low - frequency quantum correlations from a four - wave - mixing amplifier , '' * 78 * , 043816 ( 2008 ) .
v. boyer , a. m. marino , and p. d. lett , `` generation of spatially broadband twin beams for quantum imaging , '' * 100 * , 143601 ( 2008 ) .
v. boyer , a. m. marino , r. c. pooser , and p. d. lett , `` entangled images from four - wave mixing , '' science * 321 * , 544547 ( 2008 ) .
v. boyer , c. f. mccormick , e. arimondo , and p. d. lett , `` ultraslow propagation of matched pulses by four - wave mixing in an atomic vapor , '' * 99 * , 143601 ( 2007 ) .
r. c. pooser , a. m. marino , v. boyer , k. m. jones , and p. d. lett , `` low - noise amplification of a continuous - variable quantum state , '' * 103 * , 010501 ( 2009 ) .
q. glorieux , r. dubessy , s. guibal , l. guidoni , j .-
p .
likforman , t. coudreau , and e. arimondo , `` double-@xmath1 microscopic model for entangled light generation by four - wave mixing , '' * 82 * , 033819 ( 2010 ) . c. c. gerry and p. l. knight , _ introductory quantum optics _ , ( cambridge university press , 2005 ) .
h. a. bachor and t. c. ralph , _ a guide to experiments in quantum optics _ , ( wiley - vch verlag , 2004 ) . c. m. caves , `` quantum - mechanical radiation - pressure fluctuations in an interferometer , '' * 45 * , 7579 ( 1980 ) .
r. loudon , _ the quantum theory of light _ ( oxford university press , 1983 ) , 2nd ed .
r. loudon , `` theory of noise accumulation in linear optical - amplifier chains , '' ieee j. quantum electron . *
21 * , 766773 ( 1985 ) . c. m. caves and d. d. crouch , `` quantum wideband traveling - wave analysis of a degenerate parametric amplifier , '' * 4 * , 15351545 ( 1987 ) |
the pibeta collaboration has proposed a program of precise measurements of rare @xmath5 and @xmath6 decays at the paul scherrer institute ( psi ) @xcite , with particular emphasis on the pion beta decay branching ratio , @xmath7 .
the pibeta apparatus is a large solid angle non - magnetic detector optimized for detection of photons and electrons in the energy range of 5150@xmath8mev with high efficiency , energy resolution and solid angle .
the main sensitive components of the apparatus , shown and labeled in fig .
[ fig1 ] , are : * bc , a thin forward beam counter placed approximately 4@xmath8 m upstream of the detector center , ac@xmath9 and ac@xmath10 , two cylindrical active collimators , ad , an active degrader , all made of plastic scintillator and used for beam definition ; * at , a 9-element segmented active plastic scintillator target , used to stop the beam particles while simultaneously sampling the lateral beam profile ; * mwpc@xmath9 and mwpc@xmath10 , two concentric low - mass cylindrical multi - wire proportional chambers for charged particle tracking , surrounding the active target ; * pv , a fast 20-bar segmented thin plastic scintillator hodoscope , surrounding the mwpcs , used for particle identification ; * a 240-element fast high - resolution segmented spherical pure - csi shower calorimeter surrounding the target region and tracking detectors , subtending a solid angle of @xmath11% of @xmath12 ; * cv , a set of cosmic muon plastic scintillator veto counters around the entire apparatus , not shown in fig .
[ fig1 ] . to [ fig1 ]
the detector components listed above , together with delay cables for photomultiplier tube ( pmt ) analog signals , high voltage ( hv ) supplies and cables , mwpc instrumentation and gas system , fast trigger electronics , two front end computers ( one for data acquisition , the other for slow control ) , as well as the temperature control system , are all mounted on a single platform that is moved as a single unit in and out of the experimental area .
thus , the detector can become fully operational in less than 24 hours after the support platform is precisely positioned with respect to the beam line , and electrical power and ethernet connections are made .
the building and testing of the detector components were completed in 1998 , followed by the assembly and commissioning of the full detector apparatus .
data acquisition with the pibeta detector started in the second half of 1999 , initially at a reduced pion stopping rate , as planned .
since then , the pion stopping rate was gradually increased and the experiment ran during most of the available beam period in the years 2000 and 2001 at @xmath13mhz @xmath1 stopping rate . in all important respects the detector has met its design specifications . in this paper
we report on the radiation resistance and the temporal stability of the gain , energy resolution , and detection efficiency of the most affected active elements of the pibeta detector listed above .
radiation stability of the plastic scintillator detectors used in high energy and nuclear physics experiments is one of their main characteristics , and as such has been discussed in a voluminous body of research and review papers . here
we note a review paper by g. marini et al .
@xcite and refs .
@xcite , as well as references therein .
these papers address the issue of experimentally determining and improving the radiation hardness of plastic scintillators .
radiation hardness of pure ( undoped ) csi scintillators has been reported in refs . @xcite .
the pibeta measurements are performed in the @xmath14 channel at psi @xcite . for this experiment
the beam line is operated in the high intensity , low momentum resolution mode .
correspondingly , a 114 mev / c @xmath1 beam tune has been developed with momentum spread of @xmath15% and maximum nominal @xmath1 beam intensity of @xmath16/s .
the spatial spread of the @xmath1 beam is restricted by a @xmath17 cm thick lead collimator with a 7 mm pin - hole located 3985@xmath8 mm upstream of the detector center .
beam particles are first registered in the 2@xmath8 mm thick plastic scintillator ( bc ) placed directly downstream of the collimator .
pions are subsequently slowed down in the 40@xmath8 mm long active plastic degrader ( ad ) , and stopped in the active plastic target ( at ) positioned in the center of the pibeta detector , fig .
[ fig1 ] .
we have analyzed a total of 6213 production runs , for which data were accumulated between 9 october 1999 and 11 december 2000 .
this data set comprises a total number of @xmath18 beam @xmath1 s stopped in the active target .
the @xmath19 and @xmath4 beam contaminations measured in the bc at time - of - flight spectrum are small , @xmath20% and @xmath21% , respectively .
therefore , the in - beam detectors were exposed primarily to pions , while the at counters also received significant doses from the stopped pion decay products : @xmath22 .
particle discrimination between the positrons , photons and protons detected in the csi calorimeter is accomplished using the charged particle tracking detector components , i.e. , @xmath23 and pv , the plastic veto hodoscope .
all individual detector pmt analog signals are discriminated in time - over - threshold camac modules and counted with camac scaler units read out every 10@xmath8s .
the cumulative scaler counts are updated at the end of every production run in the online database , as well as saved in a computer disk file .
the most probable , as well as the average , energy deposited in each detector element are calculated in a monte carlo ( mc ) simulation using the standard detector description code geant3 @xcite .
the geant simulation also provided the average values of radiation exposure throughout the detectors volumes .
the total energy absorbed per unit detector mass exposed to radiation comprises the received radiation dose . the absorbed radiation dose for each detector is commonly expressed in units of rad @xcite , corresponding to the energy absorption of 100erg / g .
the equivalent si unit is 1 gray equaling 100rad ( @xmath24joule / kg ) .
the absorbed radiation doses for each pibeta active detector element are calculated using the experimental and mc data on ionizing particle types , cumulative particle rates and exposed detector volumes . the pibeta detector absorbed doses are listed in tables [ tab1 ] and [ tab3 ] .
the absolute energy calibration as well as the shapes of experimental deposited energy ( adc ) spectra are well understood in the mc geant simulations .
the peaks in the adc spectra accumulated for each series of 10 runs are fitted with gaussian functions in offline analysis .
the means of the gaussian functions determine the relative detector gain factors , while the standard deviations of the lower parts of the adc spectra loosely reflect the detector energy resolutions .
the pmt bias high voltages were in general kept constant , except for the csi calorimeter pmts .
the demand hv values were set in 1v steps with an accuracy and reproducibility of @xmath251v , which corresponds to an equivalent gain change of @xmath26% .
for the 220 non - peripheral csi calorimeter detectors high voltages were adjusted automatically by the online analysis computer program on a daily basis .
firm constraints were imposed on the location of the @xmath27 energy peaks .
the latter were always forced to the normalized 67.8mev value by changing the pmt high voltages and adjusting the calorimeter software gains appropriately . for a detector equipped with an @xmath28-stage pmt operating in the linear domain , a normalized real gain change @xmath29 ,
relating two different settings 1 and 2 ( at time @xmath30 and @xmath31 , respectively ) , depends on the ratio of the software gains @xmath32 and the corresponding high voltages @xmath33 : @xmath34 this equation relates the true gain of a scintillator detector at time @xmath31 to its gain at time @xmath30 .
the forward beam counter bc is the first detector placed right after the beam - defining lead collimator .
this counter tags the beam particles that have passed through the collimator .
the central part of the beam counter is a quadratic piece of bicron bc-400 plastic scintillator @xcite with dimensions @xmath35mm@xmath36 .
the scintillator is optically coupled on all four sides to four tapered acrylic lightguides .
one lightguide is glued to the scintillator edge surface with bicron bc-600 optical cement .
the other three lightguides have an air gap coupling to the scintillator . both the scintillator and the lightguides are mounted inside a light - tight enclosure .
this detector - carrying box consists of an aluminum frame with outer dimensions @xmath37mm@xmath36 , covered by two thin aluminum windows , each @xmath38 m thick .
the al frame is attached to the lead collimator and keeps the counter position fixed .
feedthroughs at the four lateral sides of the box hold the light guides and the magnetic shield cylinders of the pmts .
the box is therefore also used as a mount for the photomultipliers , while keeping the scintillator counter light - tight and protecting it mechanically .
each of the four lightguides is air - gap coupled to a hamamatsu r7400u mini - pm tube .
the stabilized active pmt voltage divider was specially designed at the university of virginia for high rate operation , because typical counter rates are @xmath39mhz .
the four analog pmt signals are electronically summed in a nim lcr 428f linear fan - in / fan - out unit set up in the experimental beam area .
the analog pulses from the beam counter are divided by a custom - made passive splitter into two signals of equal amplitude .
one signal is discriminated in a camac discriminator module ; discriminator output is digitized in a fastbus time - to - digital converter ( tdc ) as well as counted with a camac scaler unit .
the second signal is connected to a fastbus analog - to - digital converter ( adc ) , gated with a 100@xmath8ns event trigger gate .
similar electronic logic is used with other individual pibeta detectors discussed in the following sections . in order to suppress the background caused by detector hits not associated with a @xmath5-stop event we use two active beam collimators ac@xmath9 and ac@xmath10 .
both collimators are ring - shaped and are made of a 25.4@xmath8 mm thick bicron bc-400 plastic scintillator .
ac@xmath9 , the first ( upstream ) ring counter , has an outer diameter of 120@xmath8 mm and an inner diameter of 50@xmath8 mm .
ac@xmath10 , closer to the detector center , has the outer / inner diameter of 172/90@xmath8 mm , respectively .
these dimensions were chosen such that the detectors subtend the various beam tube elements and mechanical support parts of the detector without intersecting the calculated envelope of the beam .
the deposited energy ( adc signal charge ) spectrum of the forward beam counter recorded with the @xmath5-in - beam trigger is shown in fig .
most probable energy deposition by a @xmath1 beam particle corresponds to 0.70mev , while the average absorbed energy is 0.80mev .
these values are calculated with a geant monte carlo code using a 114mev / c @xmath1 beam as an input .
the slope of the low - energy ridge of the deposited energy spectrum in bc is determined by a convolution of the vavilov probability distribution and the pmt photoelectron statistics .
since the latter deteriorates with radiation damage , the slope of the low - energy edge of the adc spectrum can be used as a rough indicator of the effects of radiation damage on detector resolution , in the absence of a sharp monoenergetic line .
we have used the same approach for the other pibeta plastic scintillator detectors discussed below : ad , at , pv . to fig .
[ fig3 ] depicts the changes in the bc detector gain and fractional energy resolution as the detector became damaged by radiation over time .
it is evident from fig .
[ fig3 ] that the gain of the bc counter decreased approximately linearly over the period of one calendar year due to the radiation exposure of 2 mrad .
the decrease in gain was @xmath40%/mrad .
the fractional energy resolution for through - going pions was also slightly degraded , changing from 13.1% to 13.9% , as shown in the bottom panel of fig .
two beam breaks in the radiation exposure , totaling 98 days and 35 days , respectively , resulted in a partial beam counter recuperation and gain increases of @xmath251015% .
the annealing effect on the counter resolution is smaller , though still measurable .
to the active degrader counter is made of bicron bc-400 plastic scintillating material in the shape of a cup whose axis is aligned with the beam .
the ad fits inside mwpc@xmath9 which has an inner clearance of 90 mm .
the ad outer surfaces are slanted and connect to four acrylic lightguides .
this geometry brings the pmts out of the inner region , at the same time ensuring that the active part of ad ( cup `` bottom '' ) covers the whole target area ( 40 mm diameter ) .
the angle and wall thickness of the outer part of the ad are chosen such that a particle traveling parallel to the beam axis transverses the same thickness of 30 mm .
the angle is kept small in order to improve light collection efficiency .
each of the four fishtail lightguides ends in a rectangular cross section of @xmath41mm@xmath42 , which matches the round window area of the hamamatsu r7400u pmt .
the tubes are mounted with an air gap coupling .
the four ad pmt analog signals are summed in a nim lcr 428f linear fan - in / fan - out unit .
to the deposited energy ( adc charge ) spectrum of the active degrader counter taken with the @xmath5-in - beam trigger is shown in fig .
the lower energy part of the deposited energy spectrum is again fitted well with a gaussian function .
most probable energy deposition of the 114mev / c incident @xmath1 beam is calculated with the geant code to be equal to 13.0mev , while the average energy absorbed is 13.8mev . during the period under analysis
the degrader counter was irradiated with a dose of 1.4mrad .
the changes in the relative counter gain and energy resolution are shown in the two panels of fig .
both variables are subject to short - term variations , mostly due to changes in the beam tune .
there is , however , an average linear trend in the scintillator gain factor corresponding to 15%/mrad decrease .
the temporal degradation in the ad energy resolution is much smaller but still noticeable . to the pibeta active target , at , is a cylindrical plastic scintillator counter 50@xmath8 mm long with a diameter of 40@xmath8 mm .
the counter is segmented into 9 elements , as shown schematically in fig .
a 7 mm central cylinder is surrounded by 4 identical tubular segments that compose an inner ring with outer 30.0@xmath8 mm . the second
, outer target ring is made of 4 tubular segments rotated by 45@xmath43 with respect to the inner ring ( cf .
[ fig1]b ) .
the 9 pieces are wrapped individually in aluminized mylar foil isolating them optically from each other .
the segments are pressed together and the whole modular assembly is wrapped with black plastic tape .
each target element is acting as an independent counter , viewed by a miniature ( 8@xmath8 mm photo - cathode ) hamamatsu r7400u photomultiplier tube via a 60@xmath8 mm long , tapered acrylic lightguide .
to the uncalibrated deposited energy ( adc ) spectrum of the central target segment @xmath44 recorded with the @xmath5-in - beam trigger is depicted in fig .
the events in which the stopping pion deposits its full energy in the target element are clearly distinguished in the high - energy peak .
the peak is well described by a gaussian form with a square - root function tail .
the geant simulation predicts a peak position at 27.6mev , and the average energy deposited in the central target segment t@xmath45 of 16.6mev .
the events with a stopping particle sharing energy between two or more target segments due to scattering or to beam divergence , fall into the low energy tail .
the same low energy tail is also populated by positrons from delayed weak pion and muon decays . in table [ tab1 ]
we give the starting and final gains and resolutions for the three representative target segments . the measured temporal variation in the gain factor and energy resolution of the central target segment over the 15 month period
is depicted in fig .
the cumulative absorbed radiation dose is calculated to be 0.5mrad .
a considerable fall - off in the gain of the t@xmath45 scintillator is evident in the figure : @xmath46%/mrad . to as in the ad counter , the energy resolution is subject to short - term variations caused by changes in the beam spot geometry .
over the analyzed period , though , the resolution was degraded by a full 80% .
this level of radiation damage makes it necessary to make available identical copies of the at detector for periodic replacement .
the radiation damaged pibeta target assembly has been replaced annually with new target detectors .
mwpc@xmath9 and mwpc@xmath10 are a pair of cylindrical multiwire proportional chambers , each with one anode wire plane along the beam direction , and two cathode strip planes in stereoscopic geometry .
a general description of the design and operation of the `` dubna''-type cylindrical chambers is given in ref .
the pibeta detector cylindrical mwpcs are described in more detail in refs .
@xcite . in this section
we analyze the chamber detection efficiencies measured with minimum ionizing particles .
the detection efficiency @xmath47 of mwpc@xmath9 for minimum ionizing particles ( mips ) can be measured using copious @xmath48 ( michel ) positrons emanating from the stopping target : @xmath49 where the @xmath50 s represent the number of michel events for which all the detectors in the parenthesis register coincident hits above their discriminator threshold .
the csi calorimeter signal is discriminated with a low threshold level ( @xmath51mev ) , while a software cut on the pv deposited energy spectrum selects mip events ( fig .
[ fig8 ] ) . an equivalent expression , obtained by substituting 2 for the index 1
, holds for mwpc@xmath10 .
the average mwpc detection efficiencies at the @xmath52mhz @xmath1 beam stopping rate are @xmath53% and @xmath54% for the inner and outer chambers , respectively .
offline analysis of the individual chamber efficiencies for the period under consideration , which involved @xmath55 ( mostly minimum ionizing ) particles , found no degradation within the measurement uncertainty ( table [ tab3 ] ) .
the pv ( plastic veto ) hodoscope is located inside the csi calorimeter and surrounds the two concentric wire chambers , as shown in fig .
the hodoscope array consists of 20 thin independent plastic scintillator bars that are arranged to form a complete cylinder with a @xmath56 mm radius and a 598@xmath8 mm long axis that coincides with the beam line and the target axis .
the pv hodoscope covers the entire geometrical solid angle subtended by the csi calorimeter as seen from the target center .
a single pv detector element consists of four main components : ( 1 ) thin plastic scintillator bar , ( 2 ) two lightguides , one at each end , ( 3 ) two photomultiplier tubes , one per lightguide , and ( 4 ) aluminized mylar wrapping .
the dimensions of the individual plastic bars made of bc-400 scintillator are @xmath57mm@xmath36 .
each thin plastic scintillator bar , along with the attached lightguides , is wrapped in 0.25@xmath58 m thick aluminized mylar foil to optically separate it from the adjacent bars , and to provide a reflective surface that increases the amount of light reaching the phototubes .
scintillation light is viewed by two burle industries s83062e photomultiplier tubes , one on each end .
these tubes are 28 mm head - on fast pmts with 10 stages .
the hodoscope array is supported from the inside by a @xmath59 mm long carbon fiber cylinder with a total thickness of @xmath60 mm , corresponding to @xmath61 radiation lengths .
the hodoscope modules are kept tight around the support cylinder by a helix - wound , thin plastic string tensioned at the extreme ends of the detector stand .
the pv charged particle detection efficiency is evaluated separately for minimum - ionizing positrons with total energy above 5@xmath8mev and for non - relativistic protons with kinetic energy in the range 10 - 150@xmath8mev .
[ fig8 ] shows the measured energy spectrum of positrons and protons in the pv detector , corrected for the angle of incidence . to at the nominal @xmath1 stopping rate the individual plastic scintillator phototubes are counting at the rate of @xmath62khz , while the entire pv hodoscope is taking @xmath63 m hits / sec . the total radiation dose absorbed by the pv hodoscope array throughout the analyzed period is 40krad .
a charged particle track is defined by the coincident hits in the active target , at , two wire chambers , mwpc@xmath9 and mwpc@xmath10 , and the csi calorimeter , csi .
the pv detection efficiency is defined as the following ratio : @xmath64 where each @xmath50 represents the number of events satisfying the condition in the parenthesis .
the average pv detection efficiency measured during the detector commissioning period was @xmath65% ( table [ tab3 ] ) .
no significant long term change in the pv detection efficiency was observed . as for the previously discussed detectors , table [ tab1 ] lists the starting and final gains and resolutions for two representative pv detectors .
the temporal dependence of the detector light output for pv bar number 1 is shown in fig .
its gain factor dropped by 0.25%/krad , giving a cumulative degradation over the production period of 10% .
the energy resolution on the other hand shows a barely measurable change .
to the charged particle tracking system , combining the information from the pv hodoscope and the mwpc pair achieves mip tracking inefficiency in the range @xmath66 while operating in the environment of @xmath13mhz stopped @xmath1 s in the target .
the shower calorimeter lies at the heart of the pibeta detector .
pure ( undoped ) csi single crystal ( refs .
@xcite ) was chosen as the calorimeter material .
the pibeta calorimeter consists of 240 pure csi crystals .
spherical geometry is obtained by the ten - frequency class ii geodesic triangulation of an icosahedron @xcite .
the chosen geodesic division results in 220 hexagonal and pentagonal truncated pyramids covering a total solid angle of @xmath67sr .
additional 20 crystals surround two detector openings for the beam entry and detector access and act as electromagnetic shower leakage vetoes .
the inner radius of the crystal ball is 260@xmath8 mm , and the axial module length is 220@xmath8 mm , corresponding to 12 radiation lengths , as @xmath68 mm @xcite .
there are nine different detector module shapes : four irregular hexagonal truncated pyramids ( labeled hex - a , hex - b , hex - c , and hex - d , respectively ) , one regular pentagonal ( pent ) and two irregular half - hexagonal truncated pyramids ( hex - d1 and hex - d2 ) , plus two trapezohedrons which function as calorimeter vetoes ( vet-1 and vet-2 ) .
the volumes of the pibeta csi detector modules vary from 797@xmath8cm@xmath36 ( hex - d1/2 ) to 1718@xmath8cm@xmath36 ( hex - c ) .
the first 25 pibeta crystals were manufactured in the bicron corporation facility in newbury , ohio @xcite .
the remaining 215 csi scintillators were grown in the institute for single crystals in harkov ( amcrys ) , ukraine .
preliminary quality control , including the optical and mechanical crystal properties , was performed at the production sites . after completing physical measurements ,
all crystal surfaces were hand - polished with a mixture of 0.2@xmath6 m aluminum oxide powder and etylenglycol .
the surfaces of each csi crystal were then painted with a special organo - silicon mixture . the wavelength - shifting lacquer developed by the harkov single crystals research institute @xcite provides an optical treatment of crystal surfaces superior to more common matting or wrapping treatments .
the lacquer contains a wavelength - shifting ladder organosilicon copolymer with the chemical composition ppo+popop+coum.1 , where ppo is 2.5diphenyloxazole , popop represents 1.4-di-2-(5-phenyloxazolile - benzene ) and coum.1 is 7-diethylamine-4-methylcoumarin @xcite .
emi 9822qkb 10-stage fast photomultipliers @xcite with 75 mm end windows are attached to the back faces of hexagonal and pentagonal csi crystals using a 300@xmath6 m layer of silicone sylgard 184 elastomer ( dow corning rtv silicone rubber plus catalyst ) .
the resulting crystal photomultiplier tube couplings are strong and permanent , but can be broken by application of a substantial tangential force .
smaller half - hexagonal and trapezial detector modules are equipped with 46 mm 10-stage emi 9211qka phototubes @xcite .
both photocathodes have quartz windows transmitting photons down to @xmath70 nm . the window transparency , peaking at @xmath71 nm @xcite , is approximately matched to the spectral excitation of a pure csi fast scintillation light component with a peak emission at @xmath72 nm at room temperature @xcite .
calibrated energy spectra of the calorimeter detector show that the individual modules receive between 5 and 10mev energy depositions in a large sample of events averaged over all physics triggers .
the accumulated radiation doses for individual csi counters have been between 40rad and 160rad , table [ tab1 ] .
a deposited energy ( adc ) spectrum , calibrated in mev and representing a sum of adc values for one csi detector and its nearest neighbors , is shown as an example in fig .
[ fig10 ] .
the event trigger is a high threshold ( @xmath73mev ) delayed @xmath1 gate trigger that preferentially selects both @xmath74 positrons which have a continuous energy spectrum with a 52.5mev endpoint , and the monoenergetic 70mev @xmath75 positrons .
the @xmath6 decay positron spectrum is clipped on the left very near its endpoint by the @xmath73mev trigger discriminator threshold .
the monoenergetic @xmath5 decay positron line is used for the automatic , computer - controlled gain matching of the individual csi detectors . to
because both the software gain factors and the pmt high voltages of all csi counters are varied after each run , the relevant real gain factors are defined by eq .
( [ eq : gain ] ) .
representative gain variations for three different csi detectors are shown in figs .
[ fig11][fig13 ] .
these measurements show different gain degradation rates of up to @xmath76%/rad .
an increase in gain following extended beam breaks is a common feature of the csi light response , as seen in the three figures . to to to the overall csi calorimeter gain and energy resolution
are determined from calibrated experimental energy spectra of all 240 csi detectors .
the results are summarized in table [ tab1 ] . as it accumulated an average radiation dose of 119rad
the pibeta calorimeter gain decreased 17% and the overall energy resolution was lowered from 5.5% to 6.0% .
this decrease , however , should not be attributed entirely to radiation damage , as the crystals were subject to other sources of performance degradation over time : absorption of humidity on the crystal surfaces , slow plastic deformation of individual modules under pressure ( csi is highly malleable , while the detectors are self - supporting , with the entire weight of over 1ton distributed among the elements ) .
thus , the above performance degradation provides a very loose upper limit on the radiation effects on pure csi .
we have analyzed the temporal changes in gains , energy resolutions and detection efficiencies of the active elements of the pibeta detector during 15 month period of production data taking .
we find measurable decreases in energy gains and degradations in energy resolutions , in particular for the beam detectors that were exposed to high radiation doses as well as for more radiation - sensitive pure csi calorimeter modules .
these changes in the detector responses are monitored online throughout the production running and documented in the replay data analysis .
the changes affect the energy calibration of the pibeta detector elements and have to be properly taken into account when defining the software energy cuts in the physics analysis of rare pion and muon decays .
* acknowledgements * this work has been supported by grants from the us national science foundation .
99 d. poani et al . ,
, g. marini , i. donatelli , a. nigro , g. martellotti , m. de vincenzi , a. sciubba , f. coninckx , g. crosetti , and h. schnbacher , _ radiation damage to organic scintillation materials _ , cern yellow report 85 - 08 ( cern , geneva , 1985 ) .
a. d. bross and a. pla - dalmau , _ radiation - induced hidden absorption affects in polystyrene base plastic scintillator _ , fermilab - pub-90/224 ( fermilab , batavia , 1990 ) .
b. bicken , u. holm , t. marckmann , k. wick , and m. rohde , ieee trans .
* ns-38 * ( 1991 ) 188 .
radecs 93 conference proceedings , ieee trans
. nucl .
* ns-41 * ( 1994 ) 421 - 628 .
radecs 95 conference proceedings , ieee trans .
ns-43 * ( 1996 ) 769 - 1080 . v. g. senchishin , f. markley , v. n. lebedev , v. e. kovtun , v. s. koba , a. v. kuznichenko , v. d. tizkaja , j. a. budagov , g. bellettini , v. p. seminozhenko , i. i. zalubovsky , and i. e. chirikov - zorin , nucl .
instrum .
a * 364 * ( 1995 ) 253 .
m. l. andrieux , _ review of radiation tests on opto - electronic components and material for lhc _ , accessible at http://isnwww.in2p3.fr/atlas/andrieux/radpbn.html .
m. kobayashi , m. ieiri , k. kondo , t. miura , h. noumi , m. numajiri , y. oki , t. suzuki , m. takasaki , k. tanaka , y. yamanoi , s. sugimoto , and m. ishii , ( chamonix , france , 1992 ) , p. 513 .
a. roodman , _ crystal 2000 : proceedings of international workshop on heavy scintillators for scientific and industrial applications _
( chamonix , france , 1992 ) p. 479 . c. l. woody , j. a. kierstead , p. w. levy , and s. stoll , ieee trans .
* ns-39 * ( 1992 ) 524 .
g. eigen , and d. g. hitlin , _ crystal 2000 : proceedings of international workshop on heavy scintillators for scientific and industrial applications _
( chamonix , france , 1992 ) p. 467 .
m. kobayashi , m. ieiri , k. kondo , t. miura , h. noumi , m. numajiri , y. oki , t. suzuki , m. takasaki , k. tanaka , y. yamanoi , s. sugimoto , and m. ishii , nucl .
instrum .
meth . a * 328 * ( 1993 ) 501 .
psi users guide : accelerators facilities
( paul scherrer institute , villigen psi , 1994 ) .
r. brun , f. bruyant , m. maire , a. c. mcpherson and p. zanarini , geant3.21 dd / ee/94 - 1 ( cern , geneva , 1994 ) .
w. r. leo , _ techniques for nuclear and particle physics experiments _ ( springer - verlag , berlin , 1987 ) . bicron corporation catalog ( bicron corporation , newbury , 1989 ) , and online at url http://www.crismatec.com/. n. p.
kravchuk , phys .
particles nuclei * 25 * ( 1994 ) 1244 .
v. v. karpukhin , i. v. kisel , a. s. korenchenko , s. m. korenchenko , n. p. kravchuk , n. a. kuchinsky , n. v. khomutov , and s. ritt , nucl .
instrum .
a * 418 * ( 1998 ) 306 .
v. v. karpukhin , i. v. kisel , a. s. korenchenko , s. m. korenchenko , n. p. kravchuk , n. a. kuchinsky , n. v. khomutov , and s. ritt , instrum .
exp . tech .
* 42 * ( 1999 ) 335 .
h. kobayashi , a. konaka , k. miyake , t. t. nakamura , t. nomura , n. sasao , t. yamashita , s. sakuragi , and s. hashimoto , kyoto university preprint kuns-900 ( 1987 ) .
s. kubota , h. murakami , j. z. ruan , n. iwasa , s. sakuragi , and s. hashimoto , nucl .
inst . and meth .
a * 273 * ( 1988 ) 645 .
s. kubota , s. sakuragi , s. hashimoto and j. z. ruan , nucl .
inst . and meth .
a * 268 * ( 1988 ) 275 .
h. kenner , _ geodesic math and how to use it _ ( university of california press , berkeley , 1976 ) . , phys .
d * 66 * ( 2002 ) 01001 ; latest www update accessible at http://www-pdg.lbl.gov .
v. i. goriletsky , l. a. andryuschenko , a. m. kudin , private communication , alkali halide crystal division , institute for single crystals , national academy of sciences of ukraine ( 60 lenin prosp .
, 310001 harkov , ukraine ) .
d. r. lide , _ crc handbook of chemistry and physics _ , 71st ed . , p. 3
- 357 ( crc press , boca raton , 1990 )
. thorn - emi electron tubes catalog : photomultipliers and accessories , rockaway , nj ( 1993 ) . c. l. woody ,
p. w. levy , j. a. kierstead , t. skwarnicki , z. sobolewski , m. goldberg , n. horwitz , p. souder and d. f. anderson , ieee trans .
* ns-37 * ( 1990 ) 492 .
cccccc + detector & radiation & starting & final & starting & final + & dose ( krads ) & gain & gain & res .
( % ) & res .
( % ) + + bc & 2000 & 1.00 & 0.66 & 13.1 & 13.9 + ad & 1400 & 1.00 & 0.75 & 7.8 & 8.5 + at@xmath45 & 560 & 1.00 & 0.76 & 7.2 & 10.8 + at@xmath9 & 630 & 1.00 & 0.89 & 8.4 & 8.6 + at@xmath77 & 150 & 1.00 & 0.74 & 7.7 & 8.9 + pv@xmath45 & 44 & 1.00 & 0.99 & 31.5 & 32.1 + pv@xmath9 & 40 & 1.00 & 0.99 & 27.6 & 30.4 + pveto & 41 & 1.00 & 0.95 & 26.2 & 27.9 + csi@xmath45 & 0.046 & 1.00 & 0.89 & 5.1 & 5.5 + csi@xmath10 & 0.045 & 1.00 & 0.92 & 4.9 & 5.2 + csi@xmath78 & 0.117 & 1.00 & 1.02 & 6.0 & 6.1 + csi@xmath79 & 0.093 & 1.00 & 0.65 & 5.0 & 5.3 + csi@xmath80 & 0.152 & 1.00 & 0.86 & 6.0 & 6.5 + csi@xmath81 & 0.126 & 1.00 & 0.74 & 5.4 & 5.8 + calo & 0.119 & 1.00 & 0.83 & 5.5 & 6.0 + | we have examined long term changes in signal amplitude gain , energy resolution and detection efficiency for the active components of the pibeta detector system .
beam defining plastic scintillation counters were operated in a @xmath01mhz stopped @xmath1 beam for a period of 297 days , accumulating radiation doses of up to @xmath2rad .
detectors in the charged particle tracking system a pair of cylindrical multi - wire proportional chambers and a thin plastic scintillation barrel - shaped hodoscope array were irradiated during the same running period with an average dose of @xmath3rad . individual csi(undoped crystal )
calorimeter detectors received an average dose of @xmath0120rad , mainly from photons , positrons and protons originating from @xmath1 hadronic interactions as well as from @xmath1 and @xmath4 weak decays at rest in the active target .
* keywords : * long term temporal stability of detector gain , energy resolution and detection efficiency , radiation hardness , radiation resistance , radiation damage . |
various candidates for realizing building blocks of quantum information processors with nanoscale solid state structures have been proposed and also partially realized in ground - breaking experiments .
an important class of proposals consists of using the charge degree of freedom in semiconducting double quantum dots ( dqds ) @xcite to realize a quantum mechanical two - state system or quantum bit ( qubit).@xcite thereby , the logical states @xmath1 and @xmath2 are given by the low - energy charge states of the dqd with one excess electron either on the left or the right dot .
the transition between these states occurs via tunneling between the two dots .
a significant advantage of the charge qubit is its direct controllability via external voltage sources . in recent experiments ,
the coherent manipulation of charge states of dqds has been achieved.@xcite thereby , the control over the bias and the gate voltages permits to reliably tune the dqd in the coulomb blockade regime to the required transition line @xmath3 .
the large charging energies suppress leakage to energetically higher lying many - electron states . a central impediment for coherent quantum information processes is decoherence and dissipation , see ref .
for a particular example .
since charge qubits are rather easily addressable from outside , they are , in turn , also rather fragile for various decoherence mechanisms from the environment . in order to achieve a thorough understanding of the role of the various decoherence mechanisms
, one has to rely on realistic and precise model calculations which allow to sort out the different contributions . among them , decoherence due to acoustic phonons is one possible candidate .
this mechanism has been studied previously upon using approaches relying on the born - markov approximation.@xcite the decoherence rates obtained in these studies were one or two orders of magnitude smaller than those determined experimentally for gaas dqds.@xcite however , no exact treatment was available to validate the born - markov approximation in the dqd context .
this is particularly interesting in view of realistic dqd designs since geometrical form factors tailor a rather peculiar spectral density @xmath4 of the phonon environment . for piezoelectric phonons
, it has a high - frequency tail decaying only algebraically with @xcite @xmath5 while it is superohmic at low frequencies with @xmath6 , with a crossover at frequencies in the regime of the tunneling amplitude of the dqd . as it has also been pointed out in ref .
, the use of the born - markov approximation is appropriate only at small tunneling amplitudes .
however , is is expected to become increasingly unreliable at dqd with larger interdot tunneling amplitudes . the motivation to find exact reference solutions for this problems also stems from the fact that the born - markov approximation has led to the conclusion that other mechanisms of decoherence , as for instance , background charge fluctuations and electromagnetic noise from voltage fluctuations , would be the dominating source of decoherence while phonon decoherence should be a negligible contribution . in view of further design optimization for real devices ,
exact results are clearly desirable . in this paper , we study phonon decoherence using the same model as in ref . as a basis for the numerically exact iterative quasiadiabatic propagator path integral ( quapi ) scheme .
@xcite in particular , we avoid the born - markov approximation which turns out to be problematic at larger tunneling amplitudes @xmath7 of the dqd
. the decaying oscillations of dqd charge states allows us to extract the quality ( or q- ) factor .
we find that @xmath0 decreases with increasing tunneling amplitudes @xmath7 for small @xmath7 .
in fact , the solution in this regime is accurately described by the weak - coupling result for a superohmic environment.@xcite at large @xmath7 , the @xmath0-factor increases with increasing @xmath7 and the non - markovian corrections become noticeable changing even the order of magnitude of @xmath0 . in between these two regimes , a crossover occurs which is related to the geometry of the dqd , see below . in particular , our numerical precise method allows to obtain results for the dqd device recently realized in experiments.@xcite we find that the calculated value lies approximately two orders of magnitude above the experimentally measured value .
indeed , this suggests that the phonon decoherence mechanism is subordinate and other mechanisms dominate decoherence in dqd charge states which have been realized until now . in comparison to the results from the born - markov approximation , we find that the exact @xmath0-factor is in general larger , indicating that the former overestimates phonon decoherence .
in order to solely investigate the role of the phonons in the decoherence processes , we start with the simplifying assumption that the dqd is isolated from the leads . the total hamiltonian is given by@xcite @xmath8 here , @xmath9 is the hamiltonian of the dqd which is modeled in the basis of the two localized charge states @xmath10 and @xmath11 as a symmetric two - level - system in pseudo - spin notation ( @xmath12 are the pauli matrices ) as @xmath13 in this simple model , we have assumed that the external gate voltage is tuned such that the two charge states are energetically nearly degenerate and that the excess electron can tunnel between the two dots with the tunneling amplitude @xmath7 . a possible energy bias between the two states
could easily be included in the model but is not considered in this work .
the hamiltonian for the phonon bath is , as usual , given by @xmath14 with the phonon dispersion relation @xmath15 .
the interaction part of the hamiltonian is determined by the electron - phonon interaction and reads @xcite @xmath16 here , @xmath17 is the number of the excess electrons on the left / right ( @xmath18l / r ) dot and @xmath19 where @xmath20 and @xmath21 are the position vectors of the two dots . as in ref .
, we have assumed that the center of the left dot is located at the origin of the coordinate system while the center of the right dot is placed on the @xmath22-axis with distance @xmath23 , see fig .
[ fig.geometry ] for a sketch .
the two dots are assumed to have an equal radius of @xmath24 and to be confined to the @xmath25-plane in the underlying 2deg .
the coupling constants @xmath26 which depend on material parameters and on the wave vector @xmath27 will be specified below . to take into account a realistic dot geometry
, we associate a form factor @xmath28 to each dot , where @xmath29 is the charge density distribution of the dot . in the following ,
the two dots are assumed to have identical form factors . with this
the coupling hamiltonian simplifies to @xcite @xmath30 with @xmath31 .
notice that the phonons can propagate in all three dimensions and that the electron - phonon - coupling is not isotropic.@xcite having specified the coupling , we can now introduce the spectral density of the phonon bath which reads @xmath32 as in ref . , we specialize to linear acoustic phonons with velocity @xmath33 ( @xmath34 m / s for gaas ) and linear dispersion @xmath35 .
moreover , we only consider coupling to longitudinal piezoelectric phonons and neglect the contribution of the deformation potential which is justifiable at temperatures below 10 k for bulk gaas material.@xcite this yields the coupling constant @xmath36 , where @xmath37 is the dimensionless piezoelectric constant ( @xmath38 for gaas ) and @xmath39 being the volume of the unit cell . assuming that the charge density distribution on the dot is gaussian in the @xmath25-plane and localized in the @xmath40-direction , one finds for the spectral density of the bath @xcite @xmath41 \ , .
\label{specdens}\end{aligned}\ ] ] the spectral density is shown in the inset of fig .
[ fig.specdens ] for the case of gaas and a dot radius of @xmath42 nm . the low - frequency behavior is superohmic , i.e. , @xmath43 with @xmath44 , while the high - frequency tail falls off algebraically like @xmath45 .
the crossover between these two limiting regimes occurs on a frequency scale @xmath46 . for gaas and a typical dot radius of @xmath42 nm
, we obtain @xmath47 ghz corresponding to an energy of @xmath48ev . as we will see below , the typical tunneling amplitudes @xmath7 are comparable to this energy scale .
this indicates that the markov approximation could become problematic in this transition region .
the cut - off frequency @xmath49 corresponds to the inverse of the time scale @xmath50 of the bath autocorrelation function which reads @xcite @xmath51 .
\label{response}\end{aligned}\ ] ] the autocorrelation function @xmath52 is shown in fig.[fig.specdens ] for @xmath53 mk for @xmath42 nm and @xmath54 nm.@xcite the algebraic decay at high frequencies determines the short time behavior of @xmath52 ; notice the finite slope of @xmath55 at @xmath56 , in contrast to zero slope of the usual exponential and drude - shaped cutoff functions.@xcite the superohmic low frequency behavior of @xmath4 yields a fast decay of @xmath52 at long times .
the often used born - markov approximation corresponds to replacing the strongly peaked real part @xmath55 by a @xmath57-function with the corresponding weight while the imaginary part @xmath58 is often neglected .
moreover , we note that the dimensionless damping constant given by the prefactor @xmath59 of the low - frequency superohmic limit assumes the value @xmath60 for the parameters quoted above .
thus , our results at small @xmath7 can be compared to those from a weak - coupling approach in terms of real - time path - integrals for the spin - boson model with a superohmic environment.@xcite
the dynamics of the qubit is characterized by the time evolution of the reduced density matrix @xmath61 which is obtained after tracing out the bath degrees of freedom , i.e. , @xmath62 here , @xmath63 denotes the propagator of the full system plus bath and @xmath64 denotes the time - ordering operator .
@xmath65 is the full density operator at initial time set at @xmath56 .
we assume standard factorizing initial conditions,@xcite
i.e. , @xmath66 $ ] , where the bath is at thermal equilibrium at temperature @xmath67 and the system is prepared according to @xmath68 . throughout this work
, we always start from the qubit state latexmath:[$\rho(0 ) = @xmath61 is the well established iterative tensor multiplication algorithm derived for the quasiadiabatic propagator path integral ( quapi).@xcite it is a numerically exact algorithm , as also , for instance , the real - time quantum monte carlo method is.@xcite it has been successfully tested and adopted in various problems of open quantum systems , with and without external driving.@xcite for details of this algorithm , we refer to previous work @xcite and here only address the ingredients which are important to our calculations .
the algorithm relies on the symmetric trotter splitting of the short - time propagator @xmath70 into system ( @xmath9 ) and bath ( @xmath71 ) parts on a time slice of length @xmath72 .
the bath dynamics can be solved exactly yielding a feynman - vernon - type influence functional @xcite while the system is propagated exactly by solving the schrdinger equation .
the neglect of higher order terms of the full propagator causes an error of the order of @xmath73 .
the trotter splitting allows to calculate an approximate value of the true result for the observable of interest , with an error depending on @xmath72 . as shown in ref .
, this error vanishes proportional to @xmath74 as @xmath75 .
thus , by extrapolation of the results for @xmath75 , the trotter error can be eliminated completely ( the details of this procedure will be discussed elsewhere@xcite ) .
the interaction of the system with the bath induces internal transitions in both and creates intercorrelations between them ( memory ) ; the latter are described by the autocorrelation function @xmath52 given in eq .
( [ response ] ) . for the phonon bath , these correlations decay on a time scale given by the correlation time @xmath50 .
this motivates us to neglect such long - time correlations beyond a memory time @xmath76 and to break up the influence kernels into pieces of total length @xmath77 , where @xmath78 denotes the number of time steps over which the memory is fully taken into account .
the two strategies mentioned above are countercurrent : for the trotter splitting , a small time step @xmath72 is desirable , thus decreasing @xmath76 for a fixed @xmath78 . on the other hand ,
a large @xmath76 is wanted in order to take a long memory range into account .
nevertheless , converged results can be obtained in the following way : ( i ) we choose @xmath76 such that we include all the relevant parts of the correlation function @xmath52 .
quantitatively , this can be done by requiring @xmath79 in the asymptotic regime .
( ii ) we choose @xmath78 such that the resulting @xmath80 is small enough to ensure that we are in the regime which allows extrapolation for @xmath75 ( see above ) .
this procedure allows to completely eliminate the trotter error .
we note that the choice of @xmath78 is limited to a maximum of @xmath81 for reasonable numerical practicability on a personal computer .
we have adjusted the quapi algorithm to the phonon spectral density ( [ specdens ] ) and determine the charge population @xmath82 of the left dot as a function of time for the initial condition @xmath83 .
a typical example of the resulting damped oscillatory behavior is shown in fig .
[ fig.quapi ] where the exact quapi results are depicted by the symbols ( @xmath84 ) for the case @xmath85ev , @xmath53 mk and @xmath81 . to extract the damping rate @xmath86 and the oscillation frequency @xmath87 , we fit an exponentially decaying cosine to the numerical data .
the result of the fit is also shown in fig .
[ fig.quapi ] as solid line .
the fit yields @xmath88 for the oscillation frequency and @xmath89 for the decay constant . the ratio of @xmath86 and @xmath87 fixes the quality factor according to the convention @xmath90 .
evaluating @xmath0 allows us to investigate the dependence of the coherence of the charge oscillations on various experimental parameters .
figure [ fig.delta ] shows the @xmath0-factor as a function of the tunneling amplitude @xmath7 obtained with the numerically exact quapi algorithm ( @xmath91 ) . for illustration
, we have used parameters from the experimental setup of ref . .
since our model contains realistic assumptions about the geometry of the dqd and the materials characteristics , the predictions of the quapi calculations in this case can be considered quite accurate .
two regimes exist : ( i ) at small @xmath7 , @xmath0 decreases with increasing @xmath7 ( see below ) while ( ii ) at large @xmath7 , @xmath0 increases again with increasing @xmath7 .
the crossover in between occurs when the energy splitting between the qubit states coincides with the maximum of the spectral density @xmath4 , i.e. , when @xmath92 .
then , the damping is most efficient and thus the rate @xmath86 is maximal .
notice that realistic values for the tunneling amplitudes @xmath7 fall in the range of the bath correlation frequency scale @xmath49 ( see upper axis of fig .
[ fig.delta ] , where @xmath7 is scaled with respect to @xmath49 ) . since the coupling to the phonons is rather weak ( cf . the value of @xmath60 for the superohmic coupling constant at low frequencies ) , it is tempting to compare our exact results to simple , approximate analytical results obtained from real - time path - integral formalism in the regime of weak - coupling.@xmath93 however , we remark that one needs to have in mind that the algebraic decay of the spectral density at large frequencies is by itself not taken into account while deriving the approximate result .
thus , only the superohmic low - frequency part can be expected to yield a reasonable agreement .
the weak - coupling results are given by@xcite @xmath94^{1/2}\ ] ] and @xmath95 where @xmath96 with this , we can compute the time evolution of charge population of the left dot in the weak - coupling approximation which is shown in fig .
[ fig.quapi ] ( dashed line ) for the same parameters as above .
the deviation from the exact result for the chosen parameters due to non - markovian effects is striking and also illustrated by the numerical values @xmath97 and @xmath98 .
the resulting @xmath0-factor is shown in fig .
[ fig.delta ] ( solid line , `` wca '' ) . since the results in eqs .
( [ freq ] ) and ( [ gam ] ) are obtained by linearization with respect to the damping constant @xmath99 , they are equivalent to a born - markovian approximation .
we find noticeable deviations for intermediate and large values of @xmath7 . from eq .
( [ freq ] ) it follows that the oscillation frequency @xmath87 is renormalized by the term involving @xmath100 stemming from one - phonon interblip correlations in the self - energy.@xcite we find that at small @xmath7 these corrections are negligible , while they increasingly become important at larger @xmath7 , pointing to a non - markovian dynamics in this regime .
note that the qualitative behavior of @xmath0 versus @xmath7 is similar to that found in ref . within the born - markovian master equation approach .
however , the absolute numbers disagree .
to understand the results at small tunneling amplitudes , we zoom into the small-@xmath7 region and show the quapi results in fig .
[ fig.ohm_supohm ] ( @xmath91 , right ordinate scale ) together with the results from the weak - coupling approximation ( solid line ) obtained from eqs .
( [ freq ] ) and ( [ gam ] ) .
therefore , we have used the pure superohmic spectral density with an exponential cutoff with frequency @xmath101 , i.e , @xmath102 with @xmath103 and @xmath104 .
we find a good agreement between both results in the limit of small @xmath7 , thus justifying that the dynamics at small @xmath7 is markovian .
nevertheless , note that even in this frequency range deviations between the exact and the approximate solution do appear . as a further check
, we also compare the results of both the numerically exact and the analytical approach for @xmath0 for a pure ohmic environmental spectral density @xmath105 for @xmath106 .
both the quapi ( @xmath107 ) and the approximative weak - coupling solution ( dashed line ) coincide .
the @xmath0-factor decreases monotonically for decreasing @xmath7 since the ohmic low - frequency modes are not suppressed strongly enough in comparison with the superohmic case . although the dynamics is markovian for small @xmath7 , non - markovian corrections appear to be relevant at intermediate values of @xmath7 . because realistic values for the tunneling splitting @xmath7 are also of the order of @xmath49 , the role of the large frequency tail @xmath108 still is not negligible and can not be appropriately taken into account within a born - markovian master equation .
the sensitivity of the @xmath0-factor on the lattice temperature @xmath67 of the gaas host is shown in fig .
[ fig.temp ] for a small and a large tunneling amplitude @xmath7 .
note that for the latter the corresponding born - markov results have been shown not to be reliable due to noticeable non - markovian corrections which can be observed in our data .
another parameter which is adjustable in the dqd design is the dot radius @xmath24 .
the dependence of @xmath0 on @xmath24 is depicted in fig .
[ fig.size ] for a large and a small tunneling amplitude for a fixed ratio @xmath109 .
we find that @xmath0 decreases monotonically with increasing @xmath24 for small value of @xmath7 , while it increases with increasing @xmath24 for large value of @xmath7 , qualitatively similar to the findings in ref . .
note , however , the sizeable differences in the absolute numbers which are due to the non - markovian corrections . finally , we address the recent experiment of hayashi _
et al._.@xcite these authors implemented a gaas dqd in the bias - pulsing mode . during the pulse ,
the tunneling amplitude is constant .
the bias pulse is ramped up on a time scale of 100 ps .
the tunneling amplitude has been estimated as @xmath110 @xmath111ev .
the lattice temperature was @xmath112 mk , while the effective dot radius and the interdot distance have been estimated to be @xmath113 nm and @xmath114 nm , respectively .
for this combination of parameters , the quapi method yields @xmath115 which has to be compared with the experimental value of @xmath116 . on the other hand , the weak - coupling solution of eqs .
( [ freq ] ) and ( [ gam ] ) yields @xmath117 .
two important conclusions can be drawn from these numbers .
first , the born - markov approximation underestimates decoherence in these systems and non - markovian corrections are quite noticeable .
second , our exact result differs from the experimental value by a factor of roughly @xmath118 , indicating that the role of the phonons for the decoherence is indeed of minor importance to understand the reported experimental results .
other sources of decoherence like voltage fluctuations from the electromagnetic environment or background charge fluctuations exist and have to be taken into account for an accurate description for so far realized charge - based dqd qubits .
however , our result go beyond this point : our predicted exact values of the @xmath0-factors represent a fundamental upper limit on the possibility to improve coherence of dqd charge qubits since phonon decoherence represents an intrinsic decoherence mechanism which can hardly be overcome .
this has to be noticed also in view of the divincenzo criteria @xcite for a possible realization of a quantum computer .
to summarize , we have obtained numerically exact results for the @xmath0-factor of the decaying charge oscillations in a double quantum dot upon using the real - time quasiadiabatic propagator path integral ( quapi ) .
realistic assumptions on the form of the environmental phonon spectral density entered in our model via geometrical form factors and materials characteristics .
no fitting parameters of any sort were utilized .
we have investigated the quality ( @xmath0- ) factor as a function of the tunneling splitting and have compared our results with those obtained from a weak - coupling approximation within an analytical approach in terms of real - time path - integrals .
we find that the regime of small tunneling amplitudes is appropriately covered by the markovian description .
however , at larger ( but still realistic ) values of the tunneling amplitude , non - markovian corrections appears and are relevant .
moreover , we have determined the temperature dependence of @xmath0-factor as well as its dependence on the dot radius . from a comparison with the result obtained in an experimental realization of a double quantum dot charge qubit
, we find that the theory predicts a @xmath0-factor two orders of magnitudes larger than the measured value .
this leads to the conclusion that phonon decoherence is a subordinate mechanism in gaas quantum dots realized with present state - of - the - art technology .
clearly , other forms of coupling to the external environment prevail for those already realized devices .
however , our results also represent a fundamental upper limit to the coherence of dqd charge qubits which can hardly be overcome due to its intrinsic nature
. it would be interesting to extend the quapi calculation to si - based charge qubits , which have recently been implemented.@xcite since si has inversion symmetry , piezoelectric phonons do not occur and one needs to include deformation potential phonons in the calculations .
the decoherence induced by the electron - phonon coupling is likely to be smaller than in gaas .
however , for the deformation potential case , @xmath119 .
thus , if on one hand we expect a strong superohmic behavior at low frequencies , on the other hand , a very slow decay of the spectral function at high frequencies should occur . that may provide an even stronger non - markovian dynamics and have implications for other implementations , such as si : p donor - based charge
qubits.@xcite d. e. makarov and n. makri , chem .
lett . , 482 ( 1994 ) ; n. makri and d. e. makarov , j. chem.phys . * 102 * , 4600 ( 1995 ) ; 4611 ( 1995 ) ; n. makri , j. math.phys . *
36 * , 2430 ( 1995 ) .
m. thorwart , p. reimann , p. jung , and r. f. fox , chem .
phys . * 235 * , 61 ( 1998 ) .
m. thorwart , p. reimann , and p. hnggi , phys.rev .
e * 62 * , 5808 ( 2000 ) . | we investigate the dynamics of a double quantum dot charge qubit which is coupled to piezoelectric acoustic phonons , appropriate for gaas heterostructures . at low temperatures ,
the phonon bath induces a non - markovian dynamical behavior of the oscillations between the two charge states of the double quantum dot . upon applying the numerically exact quasiadiabatic propagator path - integral scheme ,
the reduced density matrix of the charge qubit is calculated , thereby avoiding the born - markov approximation .
this allows a systematic study of the dependence of the @xmath0-factor on the lattice temperature , on the size of the quantum dots , as well as on the interdot coupling .
we calculate the @xmath0-factor for a recently realized experimental setup and find that it is two orders of magnitudes larger than the measured value , indicating that the decoherence due to phonons is a subordinate mechanism . |
in recent years , some fundamental effects in optics have attracted much attention , such as the spin hall effect of light @xcite , optical coriolis effect @xcite , and optical magnus effect @xcite .
they manifest themselves as the splitting of photon spin states , that is , when a linearly polarized light beam passes through a refractive index gradient ( e.g. , interfaces of different media ) or an inhomogeneous medium , its left and right circular polarization components separate from each other .
this spin - dependent splitting ( sds ) effect is directly attributed to different geometrical phases that the two spin components respectively experienced , corresponding to the spin - orbital interaction @xcite .
the light beam can acquire a spin - dependent geometrical phase upon the reflection or refraction of a refractive index gradient created by the interface of different media .
when a linearly polarized paraxial light beam impinges obliquely upon this interface , the sds in real space ( coordinate space ) generates , that is , the two spin components separate from each other and reside on both sides of the incident plane @xcite .
this effect is known as the spin hall effect of light , which has recently also been extensively studied in other physical systems @xcite , in addition to optics .
actually , different refractive index gradients can give a light beam different geometrical phases , thereby resulting in specified switchable and enhanced sds effects @xcite .
therefore , the geometrical phase can serve as an alternative tool for steering photon spin states .
in general , real - space sds is often accompanied with the @xmath1-space ( momentum space ) sds , associated with spin - dependent angular shift @xcite . under normal incidence ,
both real - space and @xmath1-space splittings vanish due to the degeneracy of the geometrical phases of the two spin components @xcite . while for some inhomogeneous anisotropic media , such as surface plasmonic nanostructures , plasmonic chains , and subwavelength gratings @xcite , even as under normal incidence , the light beam can also acquire spin - dependent geometrical phases , and generates the sds in the @xmath1-space . the real - space splitting can be induced after light propagating to the far field .
some recent researches have also shown that @xmath0 plate , a uniaxial birefringent waveplate with space - variant optical axis orientations , can give a beam @xmath3 topological charge and spin - dependent geometrical phases ( @xmath0 is an integer or a semi - integer ) @xcite .
furthermore , tunable @xmath0 plates made by liquid crystals have been demonstrated to be conveniently realized , with arbitrary topological charge and geometrical phase @xcite .
therefore , from this point of view , it would be interesting to explore the sds effect in the @xmath0 plates as well as its potential ability to manipulate the photon spin states . in this work ,
we theoretically show that the @xmath0 plate can be employed to steer the sds of light in the far field .
this is due to the fact that the @xmath0 plate can apply a spin - dependent geometrical phase to a light beam that passes through it . and
the geometrical phase is tunable by varying the @xmath0 plate geometry or incident linear polarization angle .
when a linearly polarized light beam normally passes through the @xmath0 plate , @xmath1-space sds first occurs . after converging by a lens ,
the induced real - space sds in the far - field focal plane is distinguishable .
the normalized stokes parameter @xmath2 is employed to reveal the separation of spin photons , because @xmath2 represents the circular polarization degree of light @xcite .
we find that the spatial distribution of @xmath2 exhibits a multi - lobe and rotatable splitting pattern with rotational symmetry .
further , by tailoring the structure geometry of the @xmath0 plate and/or incident polarization angle of light , the lobe number , and the rotation angle both are tunable .
note that as the light beam impinges normally into the @xmath0 plate , the birefringence does not induce the separation of ordinary and extraordinary light , which will not contribute to the sds effect .
( color online ) examples of the @xmath0 plate geometries for @xmath0=1 .
the tangent to the lines shown indicates the local optical axis orientation ( fast axis ) .
( a)-(d ) represent the geometries for @xmath4=0 , @xmath5 , @xmath6 , and @xmath7 , respectively.,height=264 ]
the @xmath0 plate is essentially a uniaxial birefringent waveplate with locally varying optical axis ( fast or slow ) orientations in the transverse @xmath8 plane , having a homogeneous phase retardation @xmath9 with @xmath10 , @xmath11 , @xmath12 , and @xmath13 the refractive indices of slow and fast waves , material thickness , and operation wavelength , respectively .
it can be currently realized using liquid crystals , sub - wavelength gratings , or polymers @xcite .
the fast axis orientations , as specified by a space - variant angle @xmath14 it forms with the @xmath15-axis , are described by the following equation : @xmath16 where @xmath0 is an integer or semi - integer ; and @xmath4 indicates the angle of local optical axis direction forming with the local radial direction .
figure [ fig1 ] shows four geometries of @xmath17 for different values of @xmath4 .
the jones matrix , describing a uniaxial crystal with a fast axis in the @xmath15 direction , can be represented as @xcite @xmath18,\end{aligned}\ ] ] where @xmath19 ( @xmath20 ) is the transmission coefficient in the @xmath15(@xmath21 ) direction .
( color online ) schematic illustrating the sds produced by a @xmath0 plate .
a linearly polarized light beam normally passes through the @xmath0 plate , and is focused by a lens . @xmath22 and @xmath23 represent the left and right circular polarization components , respectively .
the two curves with arrows indicate that the splitting patterns will rotate when rotating the incident polarization direction ( @xmath24 ) .
the additional devices for measuring the stokes parameter @xmath2 are not shown.,width=321 ] as the @xmath0 plate has space - variant optical axis orientations , a position - dependent jones matrix @xmath25 can be employed to fully characterize the light propagation through the @xmath0 plate @xcite : @xmath26 where @xmath27 an input linearly polarized light , with its electric field described by a jones vector , is given as @xmath28 here , @xmath24 is the linear polarization angle of the polarization vector forming with the @xmath15-axis ; see fig .
let @xmath29 be a collimated gaussian beam : @xmath30 $ ] with @xmath31 the beam waist . for simplicity
, we neglect the absorption and loss , and let @xmath19 and @xmath20 both be equal to 1 .
then , the output electric field , @xmath32 , can be calculated as @xmath33\left(\begin{array}{ccc } 1\\ i\end{array}\right)e_0\nonumber\\ + \frac{1}{2}e^{i\theta}\left[\cos\frac{\phi}{2}+i\sin\frac{\phi}{2}e^{i\varphi}\right]\left(\begin{array}{ccc } 1\\ -i\end{array}\right)e_0 , \label{output}\end{aligned}\ ] ] where @xmath34\label{phase}\end{aligned}\ ] ] is an additional space - variant geometrical phase @xcite , depending on the local optical axis orientation of the @xmath0 plate .
@xmath35 consists of two parts , corresponding to a coherent superposition of the left and light circular polarization components , each of which is respectively composed of two terms : one carries a space - variant geometrical phase ( @xmath36 ) and @xmath3 topological charge , and the other does not .
this originates from the partial conversion of the spin - to - orbital angular momentum , with the conversion efficiency determined by @xmath37 @xcite . actually , as the incident beam
is linearly polarized , the partial left - handed photons transform into right - handed photons and acquire an additional angular momentum ( @xmath38 ) to keep the total angular momentum conserved , and vice versa @xcite . generally , the spin - orbital interaction is the origin of such fundamental effects as the spin hall effect of light , optical coriolis effect and optical magnus effect , which manifest as the sds phenomenon .
thus here , under normal incidence of a linearly polarized light , @xmath1-space sds will first occur , and then real - space sds upon propagation .
note that @xmath39 involves three free parameters , @xmath0 , @xmath4 , and @xmath24 , which serve as effective ways for manipulating the sds of light .
the momentum shift of the sds then can be calculated as @xcite @xmath40 where @xmath41 and @xmath42 , representing the left and right circular polarization components , respectively , and @xmath43 ( @xmath44 ) is the unit vector in the @xmath15(@xmath21)-direction . @xmath45 and @xmath46 are position - dependent , with the shift directions respectively determined by their signs . as the geometrical phase of the @xmath0 plate is space - variant both in the @xmath15 and @xmath21 directions ,
the @xmath1-space sds and shift in both directions occur simultaneously .
unlike here , a recent work reported a constant momentum shift in plasmonic chains where the sds effect occurred in one direction @xcite .
we then consider the far - field sds effect .
a lens with focal length @xmath47 is used here to converge the output field of the @xmath0 plate , and the field in the focal plane can be viewed as a far field , as schematically shown in fig .
[ fig2 ] . the real - space shift induced by the sds in the focal plane can be written as @xmath48 @xcite .
it is also position - dependent , which means each point in the @xmath0 plate contributes a different momentum shift ; thereby , in the far field , the final splitting effect is determined by their coherent superpositions .
the electric field distribution in the focal plane can be derived from the fraunhofer - diffraction integral formula @xcite : @xmath49}{i\lambda f_0}\int\!\!\!\int\limits_{\!\!\!\!\!\!s_{\alpha\beta } } dxdy\nonumber\\ & & \times e_{out}^{x(y)}(x , y)\exp\left[-\frac{i k_0}{f_0}(xu+yv)\right ] , \label{integral}\end{aligned}\ ] ] where @xmath50 and @xmath51 are axes parallel to the @xmath15 and @xmath21 axes , respectively , @xmath52 is the space range of the @xmath0 plate , and @xmath53 represents the @xmath15(@xmath21)-polarized component of @xmath35 . as this integral is too complicated to solve analytically , we will evaluate it numerically . to show the circular polarization degree of the resulting electric field and reveal the separation of spin photons , the stokes parameter @xmath2 is employed .
describing the sds by @xmath2 is believed to be the photonic version of a stern - gerlach experiment in the absence of a magnetic field @xcite .
the @xmath2 , normalized to the total intensity in the focal plane , can be calculated as @xcite @xmath54 where @xmath55 is the phase of @xmath56 and the denominator represents the total intensity in the focal plane .
( color online ) normalized stokes parameter @xmath2 in the focal plane under the irradiation of a linearly @xmath15-polarized light for ( a ) @xmath57 , ( b ) @xmath58 , ( c ) @xmath59 , and ( d ) @xmath60 , respectively .
other parameters are the same as in fig .
the white arrows show the clockwise rotation of @xmath2 with the increase of @xmath4.,width=321 ] we now can demonstrate the sds in the far field produced by the @xmath0 plates . as stated above , the far - field sds of left and right circular polarization components results from their spin - dependent geometrical phase experienced in the @xmath0 plate .
since the geometrical phase is space - variant , the @xmath2 parameter is inhomogeneous in the far - field focal plane .
note that @xmath39 involves three tunable parameters : @xmath0 , @xmath4 , and @xmath24 , we will then explore the influence of these parameters on the sds .
the first two parameters are associated with changing the @xmath0 plate geometry , and the third one is just related to changing the incident linear polarization angle which may be more convenient to adjust .
we first discuss the influence of the @xmath0 value on the sds .
figure [ fig3 ] shows the spatial distribution of the calculated @xmath2 parameter for @xmath61 ( @xmath57 ) under the linearly @xmath15-polarized incidence ( @xmath62 ) .
it is well known that @xmath63 corresponds to left - handed polarization helicity and @xmath64 corresponds to right - handed polarization helicity .
specifically , @xmath65 or @xmath66 corresponds to @xmath22 or @xmath23 @xcite .
one can notice that , for @xmath61 , the spatial distributions of @xmath2 show 4 , 8 , and 12 independent lobes , respectively , with alternative @xmath22 and @xmath23 components .
moreover , the @xmath2 exhibits twofold rotational ( @xmath67 ) symmetry for @xmath17 , fourfold rotational ( @xmath68 ) symmetry for @xmath69 , and sixfold rotational ( @xmath70 ) symmetry for @xmath71 , respectively .
these characteristics are determined by the rotational symmetric geometries of the @xmath0 plate and the geometrical nature of the space - variant geometrical phase [ eq .
( [ phase ] ) ] .
interestingly , a similar four - lobe spin splitting effect has also been observed in semiconductor microcavities for exciton polaritons @xcite and subwavelength metallic apertures for photons @xcite , which corresponds to the case of @xmath17 .
( color online ) normalized stokes parameter @xmath2 in the focal plane under the irradiation of a linear polarization light with polarization angle ( a ) @xmath62 , ( b ) @xmath72 , ( c ) @xmath73 , and ( d ) @xmath74 , respectively . here , @xmath57 .
other parameters are the same as in fig .
the white arrows show the counterclockwise rotation of @xmath2 with the increase of @xmath24.,width=321 ] here , we can give a qualitative explanation for this interesting sds phenomenon .
note that the output electric field of the @xmath0 plate has two parts with opposite spin handedness and topological charge ( @xmath3 ) , that is , two helical circular polarization beams [ see eq .
( [ output ] ) ] .
the wavefront of the helical beam with @xmath3 topological charge is composed of @xmath75 intertwined helical surfaces , with their handedness determined by the sign of @xmath76 @xcite . after interfering and propagating to the far - field focal plane
, these helical surfaces form @xmath77 independent lobes with alternative @xmath22 and @xmath23 components , which can be described by the @xmath2 parameter .
nonvanishing @xmath4 and @xmath24 will cause the rotation of the spatial distribution of @xmath2 , since they both can give an initial value to the geometrical phase @xmath39 .
we take @xmath17 for an example to show the influence of @xmath4 and @xmath24 on the sds . as @xmath4 and @xmath24
play similar roles in the geometrical phase with just opposite signs , their influences will cancel out each other .
when @xmath62 , increasing @xmath4 will result in the clockwise rotation of @xmath2 for the @xmath4 radian , as shown in fig .
[ fig4 ] . accordingly , increasing @xmath24 under the condition of @xmath57 will make @xmath2 rotate counterclockwise for the @xmath24 radian ( see fig . [ fig5 ] ) .
obviously , @xmath4 and @xmath24 produce the same rotation angles but opposite rotation directions .
these results confirm the predications . in the above calculations
, we have assumed that the @xmath0 plate served as quarter waveplates ( @xmath78 ) with space - variant optical axis orientations .
it is worth noting that , actually , the sds does not occur for all cases of @xmath37 . when @xmath79 ( @xmath80 is an integer ) , the term carrying the geometrical phase in eq .
( [ output ] ) vanishes , and the remaining part represents a field with the same polarization as the incident light , that is , no sds occurs .
when the @xmath0 plate serves as a half waveplate @xmath81 , @xmath82^te_0 $ ] , which represents a linearly polarized light beam with an axisymmetric polarization distribution @xcite .
thus , for these cases , the sds vanishes .
we have demonstrated a tunable sds effect in the far field produced by the inhomogeneous anisotropic media with specified geometries , called the @xmath0 plate , under the normal incidence of a linearly polarized light .
this effect originates from the spin - dependent geometrical phases of the two spin components experienced in the @xmath0 plate .
we have also shown that the sds , described by the @xmath2 parameter , exhibits a multi - lobe and rotatable splitting pattern , with the lobe number and rotation angle tunable by the geometrical phase ( associated with the @xmath0 plate geometry and the incident linear polarization angle ) . as the @xmath0 plate with arbitrary geometrical phase
could be achieved by the current fabrication technology using liquid crystals , sub - wavelength gratings , or polymers , we believe that it will serve as a potential device for manipulating the photon spin states and enables applications , such as in nano - optics and quantum information . | an inhomogeneous anisotropic medium with specific structure geometry can apply the tunable spin - dependent geometrical phase to the light passing through the medium , and thus can be used to steer the spin - dependent splitting ( sds ) of light . in this paper
, we exemplify this inference by the @xmath0 plate , an inhomogeneous anisotropic medium .
it is demonstrated that when a linearly polarized light beam normally passes through a @xmath0 plate , @xmath1-space sds first occurs , and then the real - space sds in the far - field focal plane of a converging lens is distinguishable .
interestingly , the sds , described by the normalized stokes parameter @xmath2 , shows a multi - lobe and rotatable splitting pattern with rotational symmetry .
further , by tailoring the structure geometry of the @xmath0 plate and/or the incident polarization angle of light , the lobe number and the rotation angle both are tunable .
our result suggests that the @xmath0 plate can serve as a potential device for manipulating the photon spin states and enable applications such as in nano - optics and quantum information . |
in 1882 felix klein posed the question whether every abstract riemann surface @xmath1 can be realized in euclidean 3space .
a positive answer to this question was given by a. garsia @xcite and r. redy @xcite , who proved that every riemann surface can be conformally embedded into @xmath2 .
this immediately raises the question for the optimal geometric realizations in @xmath2 of a given abstract riemann surface , or , slightly more general , for the optimal ( especially beautiful , symmetric ... ) conformal immersions of a riemann surface into space .
the same question may be raised about optimal realizations of topological types of surfaces , without the conformal constraint . in this context
, there is a general agreement that an excellent way to make the notion of `` optimal '' precise is to look for surfaces that are critical points of the willmore functional @xmath3 .
this leads to the notion of constrained willmore surfaces : an immersion @xmath4 of a riemann surface @xmath1 is a _ constrained willmore surface _
if it is critical for @xmath5 under compactly supported infinitesimal conformal variations of @xmath6 .
this generalizes the notion of _ willmore surfaces _ , the critical points of @xmath5 under all compactly supported variations . constrained willmore surfaces are a mbius invariant class of surfaces with strong links to the theory of integrable systems .
the concept of constrained willmore surfaces already appears in @xcite .
however , the euler lagrange equation characterizing constrained willmore surfaces is nowhere actually derived from the variational problem , except for the simplest cases of compact surfaces of genus @xmath7 in @xcite . examples of constrained willmore surfaces are constant mean curvature surfaces in 3dimensional space forms , see @xcite for the euclidean case and @xcite or section [ sec : constr - willm - surf ] below for general space forms . in section [ sec : constrained_variational_problems ] , we derive the euler lagrange equation for conformal immersions of compact riemann surfaces that are critical points of geometric functionals under infinitesimal conformal variations . in this euler lagrange equation , a holomorphic quadratic differential appears as a lagrange multiplier . in the case of immersions of non
compact surfaces , the same euler lagrange equation is still a sufficient ( but no longer necessary ) condition for being critical under the conformal constraint . in section [ sec : constr - minim - volume ] , we discuss examples of _ constrained minimal _ and _ constrained volume critical _ surfaces , i.e. , immersions which are critical for area or enclosed volume under the conformal constraint .
this yields valuable insights into variational problems of the type studied in section [ sec : constrained_variational_problems ] .
in particular , example [ exa : constrained_no_q ] shows that the case of open surfaces is more delicate than the compact case : we construct a constrained minimal immersion that does not satisfy the euler - lagrange equation derived in section [ sec : constrained_variational_problems ] . in section [ sec : constr - willm - surf ] , we specialize the previous results to constrained willmore surfaces and give a short proof of the fact that constant mean curvature surfaces are constrained willmore .
moreover , we describe a phenomenon first observed in @xcite : there are ( globally smooth ) immersions of compact surfaces that are constrained willmore after removing a finite number of points , but not constrained willmore as compact surfaces .
this phenomenon seems to be typical for variational problems with conformal constraint : example [ exa : discs_of_revolution ] in section [ sec : constr - minim - volume ] shows that the same can be observed in the case of constrained minimal surfaces . in the appendix
we show that for non strongly isothermic immersions of compact surfaces all infinitesimal conformal variations may be realized as derivatives of genuine conformal variations , where _ strongly isothermic _
means that the immersion admits a non trivial , holomorphic quadratic differential with the property that the null directions of its real part are principal curvature directions of the immersion . as a consequence ,
a non strongly isothermic immersion of a compact surface is a critical point of a geometric functional under infinitesimal conformal variations if and only if it is critical under genuine conformal variations .
in this section we derive the euler lagrange equation characterizing immersions of a compact riemann surface that are critical points of a geometric functional under the constraint that the conformal structure is infinitesimally preserved . in the case of
non compact surfaces , the equation thus obtained is still a sufficient condition for critical points , but ( as shown in section [ sec : constr - minim - volume ] ) it is not necessary . let @xmath1 be a riemann surface ( without boundary throughout ) and denote by @xmath8 a functional ( for non compact surfaces it is sufficient that the functional @xmath8 is well defined on compact subsets of @xmath1 ) on the space of immersions of @xmath1 into some 3dimensional riemannian manifold @xmath9 . we suppose that @xmath8 is invariant under diffeomorphisms of @xmath1 , i.e. , that it depends only on the geometry of the immersion and not on the parametrization .
moreover , we suppose that @xmath8 is differentiable in the following sense : for every immersion @xmath10 there exists a 2form @xmath11 such that the derivative of @xmath8 in direction of a compactly supported infinitesimal variation @xmath12 with @xmath13 a compactly supported function , @xmath14 a compactly supported vector field , and @xmath15 the positive unit normal , is given by @xmath16 note that @xmath17 depends only on the normal variation , because @xmath8 is invariant under diffeomorphisms . in order to derive the euler lagrange equation for variational problems with conformal constraint , we need to investigate the effect of compactly supported infinitesimal variations on the conformal structure . for this purpose
we use the fact that on oriented surfaces a conformal structure is the same as a complex structure and can be represented by an endomorphism field @xmath18 with @xmath19 . [ lem : change_of_j ] the change of the conformal structure induced by a compactly supported infinitesimal normal variation @xmath20 is @xmath21 where @xmath22 denotes the trace free part of the weingarten operator @xmath23 . for a compactly
supported infinitesimal tangential variation @xmath24 the conformal structure changes according to @xmath25 where @xmath26 denotes the lie derivative .
equation is the definition of the lie derivative . a normal variation @xmath27 changes the induced metric @xmath28
according to @xmath29 for arbitrary vector fields @xmath30 we have @xmath31 and @xmath32 .
infinitesimally , this becomes @xmath33 using the equation for @xmath34 and @xmath35 , this implies .
a compactly supported infinitesimal variation @xmath36 is called _ conformal _ if the infinitesimal change of the conformal structure induced by @xmath37 vanishes . by lemma [ lem : change_of_j ]
this is equivalent to @xmath38 clearly , compactly supported infinitesimal conformal variations can be obtained by taking the derivative @xmath39 of a genuine compactly supported conformal variation @xmath40 , i.e. , a family @xmath41 of immersions depending smoothly on @xmath42 such that on the complement of a compact subset of @xmath1 all @xmath40 coincide with @xmath6 and @xmath43 .
it should be noted that , in contrast to the case of arbitrary compactly supported infinitesimal variations , it is not true that every compactly supported infinitesimal conformal variation is the derivative of a genuine conformal variation : an example of an infinitesimal conformal variation that does not admit an extension to a genuine conformal variation is the infinitesimal variation of a planar open annulus @xmath6 given by a rotational symmetric function @xmath44 with compact support .
however , in the appendix we prove that for immersions of compact surfaces that are not strongly isothermic , every infinitesimal conformal variation is obtained from a genuine conformal variation .
a conformal immersion @xmath45 of a riemann surface into a 3dimensional riemannian manifold @xmath46 is called _ constrained @xmath8critical _ if @xmath47 for all infinitesimal conformal variations @xmath48 of @xmath6 with compact support . accordingly
, we call an immersion a _ constrained willmore surface _ , _ constrained minimal surface _ or _ constrained volume critical surface _ if it is constrained @xmath8critical for @xmath8 the willmore functional @xmath5 , the area functional @xmath49 , or the enclosed volume functional @xmath50 . for our derivation of the euler lagrange equation we need the adjoint operators to the operators assigning to a normal or tangential variation the change of the conformal structure .
we first define the adjoint of the operator @xmath51 , where @xmath52 denotes the bundle of @xmath53 anti commuting endomorphisms and where @xmath54 is the change of the complex structure induced by a normal variation @xmath55 . for this
we need the non degenerate pairing @xmath56 between 2forms @xmath57 and normal variations @xmath58 and the non degenerate pairing @xmath59 between the space of quadratic differentials @xmath60 and @xmath61 , where we use the convention that , for @xmath62 a bilinear form , the expression @xmath63 denotes the 2form @xmath64 defined by @xmath65 .
with respect to these pairings the adjoint operator @xmath66 of @xmath67 is given by @xmath68 the spaces involved in the definition of @xmath69 are visualized by the diagram @xmath70^\delta \ar@{<.>}[d ] & \gamma_0(\end_-(tm))\ar@{<.>}[d ] \\
{ \qquad \omega^2(m ) \qquad } & { \qquad \gamma(k^2),\qquad } \ar[l]^{\delta^ * } } \ ] ] where the vertical arrows indicate the pairings defined in and .
the operator @xmath71 assigning to a tangential variation the induced change of the complex structure can be interpreted as follows : under the canonical isomorphisms @xmath72 and @xmath73 , with @xmath74 the canonical bundle of the riemann surface @xmath1 , the operator @xmath71 is the usual @xmath75operator on vector fields .
we therefore denote it by @xmath76 its dual operator is as well a @xmath75operator which we denote by @xmath77 it coincides with the usual @xmath75operator on quadratic differentials . as in the case of normal variations , we visualize this by the diagram @xmath78^\dbar \ar@{<.>}[d ] & \gamma_0(\end_-(tm))\ar@{<.>}[d ] \\ { \qquad \omega^2(t^*m)\qquad } & { \qquad \gamma(k^2 ) . \qquad}\ar[l]^{\dbar^ * } } \ ] ] as an application of this language we obtain the following sufficient condition for constrained @xmath8critical immersions .
[ pro : sufficient ] let @xmath45 be a conformal immersion of a riemann surface into a 3dimensional riemannian manifold @xmath46 . if there is a holomorphic quadratic differential @xmath79 such that @xmath80 , then @xmath6 is a constrained @xmath8critical immersion .
let @xmath81 be a compactly supported infinitesimal conformal variation . by , @xmath82 and therefore @xmath83 where the last equality follows from @xmath84 .
for the case of compact surfaces the following theorem shows that this sufficient condition is also necessary .
the theorem follows essentially from the fact that , by elliptic theory of the @xmath75operator ( or the so called weyl lemma ) , for compact surfaces the space @xmath85 of holomorphic quadratic differentials is finite dimensional and @xmath86 .
let @xmath45 be a conformal immersion of a compact riemann surface
. then @xmath6 is constrained @xmath8critical if and only if there is a holomorphic quadratic differential @xmath79 such that @xmath87 [ th : euler_lagrange ] by definition , the immersion @xmath6 is constrained @xmath8critical if and only if @xmath88 for all normal deformations @xmath89 such that @xmath90 . from @xmath91
we obtain that @xmath90 is equivalent to @xmath92 .
hence , @xmath6 is constrained @xmath8critical if and only if @xmath93 .
this proves the theorem , because @xmath94 is finite dimensional and therefore @xmath95 .
we refer to @xmath96 as the _ lagrange multiplier _ of the constrained @xmath8critical immersion , because @xmath97 can be interpreted as the condition that @xmath98 is orthogonal to the space of conformal immersions @xmath99 , see the appendix for details .
we conclude the section by briefly comparing three different notions of critical points of a geometric functional @xmath100 on the space of immersions under the constraint that the conformal structure is fixed : 1 .
the immersions satisfies the euler lagrange equation @xmath101 for some lagrange multiplier @xmath79 .
the immersion is constrained @xmath8critical , i.e. , it is critical under all infinitesimal conformal variations with compact support .
3 . the immersion is @xmath8critical under all genuine conformal variations with compact support .
proposition [ pro : sufficient ] shows that i ) always implies ii ) and , by theorem [ th : euler_lagrange ] , if the underlying surface is compact i ) and ii ) are equivalent ( we will see in example [ exa : constrained_no_q ] below that this is not the case for non compact surfaces ) . by definition ,
ii ) always implies iii ) .
corollary [ cor : non_isothermic ] in the appendix shows that ii ) and iii ) are equivalent for immersions of compact surfaces that are not strongly isothermic .
in this section we discuss examples of _ constrained minimal surfaces _ and _ constrained volume critical surfaces _ , the critical points of area @xmath49 and enclosed volume @xmath50 under compactly supported infinitesimal conformal variations .
these examples reveal several remarkable properties of solutions to variational problems under the constraint that the admissible variations preserve the conformal structure : * example [ exa : spheres_of_revolution ] shows that there are immersions of compact surfaces that are not constrained @xmath100critical , but become constrained @xmath100critical after removing a finite number of points . * example [ exa : cylinder ] and [ exa : hopf_cylinders ]
show that constrained minimal immersions are not necessarily analytic .
this reflects the fact that the lagrange multiplier appearing in the euler lagrange equation for constrained minimal surfaces destroys its ellipticity .
* example [ exa : constrained_no_q ] shows that , if the underlying riemann surface is non compact , constrained @xmath100critical immersions do not necessarily admit a lagrange multiplier @xmath102 such that @xmath103 .
[ l : area_volume_gradient ] let @xmath99 be an immersion of a riemann surface @xmath1 into a 3dimensional riemannian manifold @xmath46 . under the compactly supported infinitesimal normal variation @xmath27 , the area and volume functional change according to @xmath104 where @xmath105 denotes the mean curvature and @xmath106 the area element of the immersion .
equivalently , the gradients of the functionals are the 2forms @xmath107 a proof of this lemma is left to the reader .
when considering the functionals @xmath49 and @xmath50 without the conformal constraint they behave rather differently : while @xmath50 has no critical points at all , the critical points of the functional @xmath49 are the minimal immersions into @xmath46
. in the context of the isoperimetric problem one is interested in the critical points of @xmath49 under the constraint that @xmath50 is fixed ( or vice versa ) .
solutions to this constrained variational problem are the surfaces of constant mean curvature , since @xmath6 is critical under the constraint if and only if @xmath108 . in euclidean space there
exist no immersions of compact riemann surfaces that are constrained minimal or constrained volume critical , because homotheties do not change the conformal structure but scale @xmath49 and @xmath50 .
in contrast to this , there are compact constrained minimal and constrained volume critical surfaces in the 3sphere . _
every non totally umbilic homogeneous surface in a space form of dimension @xmath109 is constrained minimal and constrained volume critical .
_ this can be seen as follows . because homogeneous surfaces have constant mean curvature
, the codazzi equation implies that the hopf differential , i.e. , the unique quadratic differential @xmath110 that satisfies @xmath111 , where @xmath112 denotes the trace free second fundamental form , is holomorphic . from @xmath113 , @xmath114 and @xmath115 follows that the hopf differential satisfies @xmath116 where @xmath117 denotes the gaussian curvature , i.e. , @xmath118 . since @xmath119 is a nonzero constant , there are @xmath120 and @xmath121 such that @xmath122 and @xmath123 , which , by proposition [ pro : sufficient ] , proves the statement .
the non totally umbilic homogeneous surfaces in space forms are isometric to the following examples : the tori @xmath124 with @xmath125 in @xmath126 , the cylinders @xmath127 in @xmath2 and the cones in the half space model of hyperbolic 3space whose cusp lies on the infinity boundary and whose axis of rotation is perpendicular to the infinity boundary .
the exclusion of totally umbilic surfaces in the above statement is necessary : every piece of the plane is minimal but not constrained volume critical , because every infinitesimal deformation with compact support is conformal .
for the same reason , every piece of the round sphere is neither constrained minimal nor constrained volume critical .
[ exa : discs_of_revolution ] let @xmath128 be a conformal immersion of the open unit disc into @xmath2 that , on the punctured disc @xmath129 , parametrizes a surface of revolution .
then every compactly supported variation of @xmath6 that preserves the rotational symmetry is conformal
. 4 cm + + figure 1 .
deformation of surfaces of revolution .
+ the reason is that for immersions of the disc with rotational symmetry , the meridian curve has infinite length in the hyperbolic half plane whose boundary is the axis of rotation , and , moreover , such immersions are conformal if and only if , with respect to polar coordinates @xmath130 , the meridian curve is parametrized by hyperbolic arc length .
thus , deforming the profile curve one can decrease @xmath49 and @xmath50 and therefore : _ an immersion of a disc with rotational symmetry is never constrained minimal nor constrained volume critical , unless it is a planar disc and therefore minimal . _
[ exa : spheres_of_revolution ] figure 2 shows four constrained minimal surfaces with rotational symmetry @xcite .
these surfaces can be smoothly extended to compact embedded spheres which are not constrained minimal .
the preceding argument shows that only after removing both points on the axis of revolution one obtains constrained minimal surfaces .
figure 2 . constrained minimal surfaces .
interestingly , the round sphere with poles removed can be approximated by such punctured constrained minimal spheres with rotational symmetry although , by example [ exa : discs_of_revolution ] , a piece of the sphere itself is never constrained minimal .
[ exa : cylinder ] this example shows that , in contrast to minimal surfaces , constrained minimal surfaces are not necessarily analytic .
_ every cylinder over a plane curve is constrained minimal . _ to see this , assume that the plane curve @xmath131 is parametrized with respect to arc length .
then the cylinder @xmath132 is an isometric immersions of the plane with weingarten operator @xmath133 and mean curvature @xmath134 where @xmath135 denotes the curvature of @xmath136 . for the holomorphic quadratic differential @xmath137 with @xmath138 we have @xmath139 and therefore @xmath140 which
, by proposition [ pro : sufficient ] , proves the statement .
[ exa : constrained_no_q ] we give now an example of a constrained minimal immersion that does not admit a holomorphic quadratic differential @xmath96 with @xmath141 .
this shows that theorem [ th : euler_lagrange ] does not hold in the non compact case .
the example is constructed as follows : take a planar domain as in figure 3 and bend the hatched areas upwards such that all 3 fingers of the surface are cylinders over plane curves lying in three different planes perpendicular to the article .
as shown in example [ exa : cylinder ] the surface thus obtained is locally constrained minimal and locally admits a holomorphic quadratic differential @xmath96 such that the euler lagrange equation @xmath141 is satisfied .
it remains to be proven that the surface is globally constrained minimal but does not admit a global holomorphic quadratic differential @xmath96 such that @xmath141 .
4.4 cm to prove the first claim , we have to check that @xmath142 for every compactly supported infinitesimal conformal variation @xmath143 . on the white planar area
@xmath144 vanishes and @xmath145 vanishes for all normal variations @xmath146 .
hence , without loss of generality we may assume that @xmath147 has support in a small neighborhood of the hatched areas : changing @xmath147 on the white domain to achieve this would neither change the integral @xmath148 nor would it destroy conformality of the variation . but this proves @xmath149 , because the 3 fingers of the surface are constrained minimal by example [ exa : cylinder ] . + the fact that there is no global holomorphic quadratic differential @xmath96 such that @xmath141 can be seen as follows : as in example [ exa : cylinder ] , denote by @xmath150 , @xmath151 the squares of the differentials of the cylinder coordinates on the three fingers . on each finger
we then have @xmath152 for the respective @xmath153 .
moreover , if there was a global holomorphic quadratic differential @xmath96 with @xmath141 , then on each finger there had to be @xmath154 such that @xmath155 , which is impossible .
[ exa : hopf_cylinders ] _ all hopf cylinders in @xmath156 , in particular all hopf tori , are constrained minimal and constrained volume
critical_. a surface in @xmath156 that is the preimage @xmath157 of an immersed curve @xmath136 in @xmath158 under the hopf fibration @xmath159 is called a hopf cylinder or , if @xmath136 is closed , a hopf torus .
taking arc length parameters of a horizontal lift of @xmath136 and of the hopf fibers yields an isometric parametrization of the hopf cylinder . with respect to
the corresponding coordinates @xmath160 , the weingarten operator takes the form @xmath161 , cf .
@xcite , and the mean curvature @xmath162 , where @xmath135 denotes the curvature of @xmath136 in @xmath158 .
the holomorphic quadratic differential @xmath137 satisfies @xmath163 and @xmath164 . by lemma [ l : area_volume_gradient ] and
proposition [ pro : sufficient ] this proves the claim .
using theorem [ th : constrained_willmore ] of the next section one gets that _ a hopf torus is constrained willmore if and only if there are constants @xmath165 such that _
@xmath166 this equation means that the generating closed curve @xmath136 is a critical point of @xmath167 for variations fixing the length and the enclosed area , i.e. , _
@xmath136 is a generalized elastic curve _ , cf .
this reflects the fact , see @xcite , that the hopf torus generated by @xmath136 has willmore energy @xmath168 and that fixing the conformal type of a hopf torus means fixing the length @xmath169 and enclosed area @xmath23 of @xmath136 , because the hopf torus is isometric to @xmath170 with @xmath171 generated by @xmath172 and @xmath173 .
the willmore energy or elastic bending energy was suggested as a global invariant of surfaces by t. willmore @xcite in 1965 , but appears already in earlier work of s. germain @xcite and w. blaschke @xcite .
its applications range from the biophysics of membranes to string theory , where it was introduced by w. helfrich and a. m. polyakov , respectively .
the willmore energy of an immersion @xmath45 into a 3dimensional riemannian manifold @xmath46 is @xmath174 where @xmath105 is the mean curvature and @xmath175 the sectional curvature of @xmath176 along @xmath177 . using the gauss equation @xmath178 which relates the sectional curvature @xmath74 of the induced metric on @xmath1 to the gaussian curvature @xmath118 of the immersion the willmore functional can be rewritten as @xmath179 where @xmath180 denotes the squared length of the trace free second fundamental form .
the functional @xmath181 is invariant under conformal transformations , because the integrand @xmath182 does not change under conformal scalings of the ambient metric . by the gauss bonnet theorem , since we are considering only compactly supported variations , being a critical point of the willmore functional is a conformally invariant property , and , for compact surfaces , the willmore functional @xmath5 itself is conformally invariant .
the willmore functional @xmath5 and the functionals @xmath183 and @xmath184 , with @xmath185 , @xmath186 denoting the principal curvatures , differ by multiples of @xmath187 only and therefore have the same critical points .
thus , the willmore functional can be seen as a measure of the total amount of principal curvature and the defect from being totally umbilic .
a conformal immersion @xmath188 of a riemann surface @xmath1 into a 3dimensional riemannian manifold @xmath189 is _ constrained willmore _ if it is a critical point of the willmore functional @xmath5 under compactly supported infinitesimal conformal variations . for simplicity
we restrict now to the case that @xmath46 is a manifold of constant sectional curvature . because both the functional and the constraint are invariant under conformal changes of the metric @xmath190 , the notion of constrained willmore surfaces for all three space forms coincides and
could as well be considered from a purely mbius geometric viewpoint , see e.g. @xcite .
a proof of the following theorem can be found in @xcite .
[ th : gradient_will ] a compactly supported infinitesimal variation @xmath36 of an immersion @xmath45 into a 3dimensional space form changes the willmore functional according to @xmath191 in particular , the gradient of the willmore functional is the 2form @xmath192 proposition [ pro : sufficient ] and theorems [ th : euler_lagrange ] and [ th : gradient_will ] imply the following theorem .
[ th : constrained_willmore ] a conformal immersion of a compact riemann surface into a 3dimensional space form is constrained willmore if and only if there exists a holomorphic quadratic differential @xmath79 such that @xmath193 for non
compact surfaces , the existence of a holomorphic quadratic differential @xmath79 that satisfies implies that the surface is constrained willmore .
it should be noted that willmore surfaces correspond to the case that @xmath194 in .
in fact , even for immersions of non compact surfaces the euler lagrange equation of willmore surfaces is a necessary condition : an immersion of a ( possibly open ) surface into a 3dimensional space form is willmore if and only if @xmath195 .
we now give a simple proof of the following result , cf .
@xcite .
[ cor : cmc_is_constrained_willmore ] every constant mean curvature surface @xmath188 in a 3dimensional space form @xmath46 is constrained willmore . for constant mean curvature surfaces ,
the gradient of @xmath5 is given by @xmath196 .
moreover , by , the holomorphic hopf differential @xmath197 satisfies @xmath198 .
this proves the statement , because the holomorphic quadratic differential @xmath199 satisfies @xmath200 in addition to being constrained willmore , constant mean curvature surfaces in space forms are isothermic , which means that away from umbilics they admit conformal curvature line coordinates .
hence , constant mean curvature surfaces belong to two important surface classes of mbius geometry : that of constrained willmore surfaces and that of isothermic surfaces . for a constrained willmore immersion of a torus into the conformal 3sphere , j. richter conversely proved ( cf .
@xcite ) that , if the immersion is also strongly isothermic as defined in the appendix , it has constant mean curvature with respect to some space form subgeometry .
this generalizes a theorem of g. thomsen @xcite : a surface is willmore and isothermic if and only if it is minimal in some space form .
in contrast to thomsen s theorem , the global assumption in richter s result is essential : an example , which is due to f. burstall , of an isothermic , constrained willmore surface that does not have constant mean curvature in some space form is the cylinder over the plane curve with @xmath201 , @xmath202 and @xmath203 , @xmath204 . ] in figure 4 .
+ figure 4 . generating curve of a burstall cylinder .
a fundamental property of constrained willmore surfaces is the existence of an associated family depending on a spectral parameter , see e.g. @xcite .
this relation to the theory of integrable systems has important consequences .
for example , in @xcite the associated family of flat connections is used in order to show that to every constrained willmore torus one can assign a compact riemann surface , the so called spectral curve ( a different approach to proving this is described in @xcite ) .
this fact allows to parameterize constrained willmore tori explicitly in terms of holomorphic functions .
the case of constrained willmore immersions of the sphere is the only case where a complete classification ( in the sense of a reduction to a simpler algebraic geometric problem ) is known : because there is only one conformal structure on the sphere , all constrained willmore spheres are willmore and one can apply r. bryant s result @xcite according to which all willmore spheres in the conformal 3sphere are complete euclidean minimal surfaces of finite total curvature with planar ends ( and therefore algebraic ) .
bryant proved that the willmore energy of willmore spheres is quantized and that the possible willmore energies are @xmath205 . from the integrable systems point of view
it is interesting that willmore spheres are examples of so called soliton spheres , see @xcite .
based on the idea of smooth ends @xcite for constant mean curvature 1 surfaces in hyperbolic 3space one can construct a family of globally smooth conformal immersions of the sphere that , by corollary [ cor : cmc_is_constrained_willmore ] , are constrained willmore after removing a finite number of points the `` ends '' at which the immersion touches the ideal boundary of hyperbolic space .
these so called bryant spheres with smooth ends are constrained willmore as punctured spheres only , but not as compact surfaces , because then they had to be willmore which is impossible by bryant s result on willmore spheres .
nevertheless , they obey the same quantization of the willmore energy as willmore spheres , i.e. , @xmath205 , and they also are soliton spheres .
figure 5 shows a bryant sphere with two smooth ends .
+ figure 5 .
catenoid cousin with smooth ends .
the phenomenon that an immersion of a surface becomes constrained willmore only after removing a finite number of points does not occur for willmore surfaces , i.e. , for the critical points of the willmore energy under all variations .
in the example of bryant spheres with smooth ends , the holomorphic quadratic differential in the euler lagrange equation has poles at the ends , which have to be removed in order to obtain a constrained willmore surface . because the left hand side in is globally smooth , the ends have to be umbilic points .
our definition of constrained @xmath8critical immersions in section [ sec : constrained_variational_problems ] demands that the immersion is critical for all _ infinitesimal _ conformal variations with compact support
. constrained @xmath8critical immersions according to this definition are in particular critical under all _ genuine _ conformal variations with compact support . in this appendix
we show that for non strongly isothermic immersions of compact surfaces it is conversely true that being critical with respect to genuine variations implies being critical for all infinitesimal conformal variations .
we call a conformal immersion @xmath188 of a riemann surface @xmath1 into a 3dimensional riemannian manifold @xmath189 _ strongly isothermic _ if there exists a non trivial holomorphic quadratic differential @xmath79 with the property that the null directions of @xmath206 are principal curvature directions of @xmath6 .
[ pro : isothermic_is_deltastar_of_q_eq_0 ] let @xmath6 be a conformal immersion of a riemann surface and let @xmath102 be a non trivial holomorphic quadratic differential .
the null directions of @xmath206 are principle curvature directions of @xmath6 if and only if @xmath207 for @xmath69 as in .
the zeros of @xmath96 are then umbilic points of @xmath6 .
away from the isolated zeros of @xmath96 one can choose local coordinates @xmath208 such that @xmath209 on @xmath210 .
then @xmath211 and we have @xmath212 where @xmath213 denotes the off diagonal term of the weingarten operator @xmath23 with respect to the coordinates @xmath214 .
thus , @xmath215 holds on @xmath210 if and only the @xmath214parameter lines are curvature lines of @xmath6 .
furthermore , it follows that the zeros of @xmath96 are umbilics of @xmath6 .
the proof of proposition [ pro : isothermic_is_deltastar_of_q_eq_0 ] shows that a strongly isothermic immersion is isothermic in the sense that , away from umbilics , it admits conformal curvature line coordinates .
examples of strongly isothermic immersions are surfaces of revolution and constant mean curvature surfaces in space forms . for the latter ,
a holomorphic quadratic differential satisfying @xmath215 is given by @xmath216 , where @xmath197 denotes the hopf differential .
let @xmath1 be a compact , oriented surface of genus @xmath217 .
the teichmller space @xmath218 of @xmath1 is the quotient of the set of complex structures @xmath219 on @xmath1 by the group of diffeomorphisms of @xmath1 that are homotopic to the identity .
it is well known , see e.g. @xcite , that @xmath218 is a finite dimensional smooth manifold and that the projection @xmath220 is differentiable .
the kernel of its differential at @xmath53 is the image of @xmath221 , i.e. , the space that , with respect to the pairing , is perpendicular to the finite dimensional vector space of holomorphic quadratic differentials of the riemann surface @xmath222 : @xmath223 an immersion @xmath188 induces a unique complex structure @xmath224 compatible with the orientation and the induced metric .
let @xmath225 denote the composition of the map @xmath226 with @xmath227 .
by lemma [ lem : change_of_j ] and , for every infinitesimal variation @xmath36 of @xmath6 , the differential @xmath228 of @xmath225 at @xmath6 is @xmath229 hence , @xmath228 is surjective if and only if @xmath230 , which , by , is equivalent to @xmath231 , because @xmath232 has finite dimension . [ the : appendix ] let @xmath188 be an immersion of a compact riemann surface . if @xmath6 is not strongly isothermic , then every infinitesimal conformal variation of @xmath6 is the derivative of a genuine conformal variation .
[ cor : non_isothermic ] an immersion @xmath188 of a compact riemann surface that is not strongly isothermic is constrained @xmath8critical if and only if it is a critical point of @xmath8 under all genuine conformal variations . by proposition [ lem : strongly_isothermic ]
, if @xmath6 is not strongly isothermic , the differential @xmath233 of the map @xmath225 at @xmath6 is surjective .
hence , there exist functions @xmath234 such that the images of the corresponding infinitesimal normal variations @xmath235 span the tangent space to teichmller space at @xmath236 .
let @xmath237 be an infinitesimal conformal variation of @xmath6 , i.e. , @xmath238 .
then , in a neighborhood of zero in @xmath239 the maps @xmath240 are immersions and @xmath241 is a submersion onto teichmller space .
the implicit function theorem implies that there exists a curve such that @xmath242 and @xmath243 . in particular , taking the derivative of @xmath244 at @xmath245 yields @xmath246 which implies @xmath247 .
hence , the genuine conformal variation @xmath248 of @xmath6 satisfies @xmath249 . | constrained willmore surfaces are conformal immersions of riemann surfaces that are critical points of the willmore energy @xmath0 under compactly supported infinitesimal conformal variations .
examples include all constant mean curvature surfaces in space forms . in this paper
we investigate more generally the critical points of arbitrary geometric functionals on the space of immersions under the constraint that the admissible variations infinitesimally preserve the conformal structure . besides constrained willmore surfaces we discuss in some detail examples of constrained minimal and volume critical surfaces , the critical points of the area and enclosed volume functional under the conformal constraint . in calc .
var .
partial differential equations * 32 * ( 2008 ) , 263 - 277 . ] |
in the last decades the study of nematic liquid crystals in the presence of microstructured substrates has been the subject of intense research @xcite .
this problem is interesting not only from a fundamental point of view , but also due to its practical applications , such as the design of zenithally bistable devices @xcite , or the trapping of colloidal particles on specified sites @xcite .
the presence of the structured substrate typically distorts the nematic orientational order , leading to elastic distortions and the formation of topological defects . on the other hand
, the substrate topography can determine the director orientation far away from the substrate .
since the seminal work of berreman @xcite , this problem has been extensively studied and generalized in the literature @xcite . wetting and filling transitions by nematics on these surfaces have also been studied @xcite . when the substrate has cusps , disclination - like singularities generally appear at or very close to them @xcite . however , even when the substrate is smooth disclination lines may appear , in the nematic , close to the substrate @xcite . in both cases , these orientation - field singularities play an important role to understand the different textures of the nematic in contact with a patterned substrate under strong anchoring conditions . in this paper
we investigate the equilibrium nematic texture at substrates of arbitrary cross section , and determine how this texture determines the ( homogeneous ) nematic director in the far - field . anchoring transitions are identified as the transitions between nematic textures with different far - field orientations .
we will assume that the nematic director field lies in the plane perpendicular to the patterned substrate longitudinal axis , and that the substrate lengthscales are large enough to ensure strong anchoring conditions on the substrate .
in a few cases analytical results are available @xcite , but in general we have to resort to numerical methods .
most studies involve numerical minimization of free - energy functionals such as the landau - de gennes or frank - oseen models .
the landau - de gennes model describes the emergence of topological defects , but it is very time - consuming and it is difficult to obtain systematic results when the pattern length scales are much larger than the nematic coherence length . on the other hand , in the frank - oseen model topological defects need to be included by hand . in ref .
@xcite we extended the frank - oseen model to include disclination - like singularities near the cusps of a sawtooth substrate , and found excellent agreement with previous landau - de gennes calculations @xcite . in this paper
we generalize this numerical method to surface reliefs of arbitrary sections ( with or without cusps ) , as well as in the presence of disclination lines in the nematic phase .
we apply our method to study the anchoring transitions induced by the patterned substrate in sawtooth , crenellated and sinusoidal substrates under strong anchoring conditions .
the paper is organized as follows . in section [ sec2 ]
is devoted we set the problem .
the numerical technique is presented in section [ sec3 ] .
results on sawtooth , crenellated and sinusoidal substrates are discussed in section [ sec4 ] .
finally , we end with the conclusions in section [ sec5 ] .
and the relief profile @xmath0 . ] we consider a nematic liquid in contact with a patterned substrate with a relief profile @xmath1 that favours local homeotropic anchoring of the molecules ( see fig . [ fig1 ] ) .
translational symmetry along the @xmath2 axis is assumed , so that @xmath3 . on the other hand ,
the substrate is periodic along the @xmath4-axis with a wavelength @xmath5 , i.e. @xmath6 .
furthermore , we assume that the nematic director field @xmath7 exhibits only in - plane distortions , and thus it can be parametrized by the angle @xmath8 between the local director and the @xmath9 axis , yielding @xmath10 . the nematic order may be represented locally by a traceless symmetric second - rank tensor order parameter @xmath11 , with cartesian components @xmath12 + \frac 1 2 b [ l_il_j - m_i m_j]$ ] , where @xmath13 is the nematic order parameter , which measures the orientational ordering along the nematic director , and @xmath14 the biaxiality order parameter , which measures the ordering of the molecules along directions perpendicular to @xmath15 , characterized by the eigenvectors @xmath16 and @xmath17 .
we consider uniaxial nematic liquid crystals , so that @xmath18 except close to the substrate or within the topological defect cores .
note that an inversion of @xmath15 does not change the value of @xmath11 , and thus in nematics , @xmath15 and @xmath19 are physically equivalent configurations . far from the substrate ,
no specific orientation is imposed , but we require that the bulk nematic phase is oriented uniformly along some direction not specified , implying that @xmath20 as @xmath21 .
we anticipate that different nematic textures close to the patterned surface will lead , in general , to distinct nematic far - field orientations , and thus surface transitions will change the nematic anchoring with respect to the substrate reference plane @xmath22 . in general different textures have different symmetries , and thus the anchoring transitions are expected be first - order .
close to the transitions the surface states are locally stable in a thermodynamic sense , and these geometries are ideal candidates for the design of zenithally bistable devices .
the excess free energy @xmath23 can be written as @xmath24 , where @xmath25 is the elastic contribution , @xmath26 is the free - energy contribution associated to the anchoring of the nematogen molecules to the substrate and @xmath27 is the contribution associated to the disclination cores .
@xmath25 is given by the frank - oseen elastic free energy @xcite : @xmath28\bigg ] \nonumber\end{aligned}\ ] ] where @xmath29 is the volume occupied by the nematic liquid crystal , @xmath30 , @xmath31 and @xmath32 are the splay , twist and bend bulk elastic constants , respectively , and @xmath33 is the saddle - splay elastic constant .
thus , using the parametrization of @xmath15 in terms of the field @xmath34 , the elastic contribution to the nematic free energy per unit length along the @xmath2 axis , @xmath35 , is : @xmath36 where @xmath37 is the @xmath38 section of the volume occupied by the nematic liquid crystal .
note that the twist and saddle - splay contributions vanish identically .
if we further assume that the splay and bend elastic constants are equal , i.e. @xmath39 , then : @xmath40 the anchoring free - energy contribution @xmath26 is modelled by the rapini - papoular approximation @xcite @xmath41 where the integral is over the substrate surface @xmath42 , and @xmath43 is the outwards unit vector normal to the substrate at @xmath44 .
thus , the anchoring contribution to the nematic free energy per unit length along the @xmath2 axis , @xmath45 , is : @xmath46 where @xmath47 is the curve @xmath48 , parametrized by its natural parameter @xmath49 , and @xmath50 is the angle between @xmath15 and @xmath51 , i.e. @xmath52 , with @xmath53 . where the equation for the orientational field @xmath8 is solved .
@xmath5 is the period of the substrate relief and @xmath54 is the height of the domain . ] in order to obtain the equilibrium nematic texture , we have to minimize the functional @xmath55=f_e+f_a+f_c$ ] , with @xmath56 and @xmath57 given by eqs .
( [ frank3 ] ) and ( [ rapini2 ] ) , respectively . due to the symmetries of the problem
, we have to find the solution only in the @xmath38 region @xmath58 delimited by the dashed line in fig .
[ fig2 ] and bounded below by one period of the substrate relief . the euler - lagrange equation associated to the functional @xmath59
$ ] reduces to the laplace equation in @xmath58 , @xmath60 , subject to the boundary condition on the substrate : @xmath61 where @xmath62 is the klman - de gennes extrapolation length @xcite .
in addition , periodic boundary conditions are imposed on the sides @xmath63 and @xmath64 , and free ( neumann ) boundary conditions @xmath65 at the far field .
some comments are in order at this point .
first , other boundary conditions on the lateral sides are allowed .
for example , we can assume that @xmath66 , where @xmath67 is an integer .
however , this condition leads to a solution of the laplace equation which is linear in @xmath4 far away from the substrate , leading to an infinite interfacial free energy as a result of elastic distortions in the bulk .
as we are interested in the anchoring of an undistorted bulk nematic at a patterned substrate , we discard these solutions . on the other hand , we take the limit @xmath68 .
finite values of @xmath54 are relevant to study nematic textures under confinement . the continuum model described above is valid as long as all the characteristic relief lengths are much larger than the nematic coherence length @xmath69 , which is of the order of the molecular size . for strong anchoring conditions , @xmath70 and eq .
( [ bc1 ] ) can be substituted by the strong anchoring condition @xmath71 on the substrate . for weak
anchoring , @xmath72 , and in principle this lengthscale is also relevant . in typical liquid crystals
@xmath73 nm and @xmath74 .
however , if we assume that the typical lengthscale which controls the substrate pattern is @xmath5 , then strong anchoring conditions on the substrate may be considered also for @xmath72 if @xmath75 .
we can justify this result by rescaling the domain by a factor @xmath5 @xcite : @xmath76 and @xmath77 . in order to minimize the free energy functional in this rescaled description , we solve the laplace equation in the rescaled domain subject to an effective anchoring potential @xmath78 on the substrate ,
so the rescaled extrapolation length is @xmath79 in this limit .
the question about the existence of anchoring transitions in the strong anchoring limit is related to the uniqueness of the solutions of the laplace equation subject to the boundary conditions described above .
a standard analysis of the mathematical problem shows that the laplace equation is unique for a given integration domain and boundary conditions on the substrate .
in fact , the eigenvalues of the laplacian with the corresponding homogeneous boundary conditions are strictly positive , as inferred from its rayleigh quotient , and the uniqueness of the solution is guaranteed via fredholm s alternative .
thus , only one texture is expected under strong anchoring conditions .
this result is consistent with previous results in the literature @xcite , where it is shown that for patterned and/or chemically heterogeneous substrates a unique nematic texture is observed if the extrapolation lengths are much larger than @xmath5 . on the other hand , a transition between an almost homogeneous and a distorted texture may be observed if any of the relevant extrapolation lengths is of order of @xmath5 @xcite .
the driving force for this transition is the competition between the elastic and the anchoring contributions to the free energy .
this mechanism is not relevant at large @xmath5 , since @xmath80 while @xmath81 when @xmath5 is large . however , there is analytical , numerical and experimental evidence that different textures are indeed possible for a given substrate relief , even in the strong anchoring limit @xcite .
the apparent contradiction with the previous result may be resolved by noting that some of the textures exhibit topological defects in the nematic , i.e. @xmath82 disclination lines or , in their absence , the relief has cusps . in the first situation ,
the presence of a disclination line in the nematic modifies the domain where the laplace equation is solved , since the solution has a singularity at the defect core .
in addition , a branch cut must be added in order to avoid the non - physical ( but mathematically sound ) jump of @xmath8 by @xmath83 when following a loop enclosing the defect core .
thus the domain where the laplace equation is solved has to be modified by excluding both the defect core and the branch cut , with additional conditions on these new boundaries .
furthermore , the free energy per unit length will have a contribution @xmath84 due to the destruction of orientational order at the core of the disclination line . on the other hand , when the substrate relief has cusps
, the angle @xmath8 exhibits a discontinuity under strong anchoring conditions .
this implies that the solution has a disclination - like singularity close to the cusp , with its associated core contribution @xmath84 @xcite .
the strength of the singularity @xmath85 , i.e. its effective topological charge , is related geometrically to the opening angle of the cusp @xmath86 and the jump @xmath87 of the orientational field @xmath8 when crossing the cusp along the surface from right to left as @xmath88 @xcite .
one posibility is that @xmath89 , which leads to @xmath90 .
however , this is not the only option due to the physical equivalence between @xmath15 and @xmath19 .
in particular , it is posible that @xmath91 for @xmath92 , which leads to @xmath93 . in each case , the boundary conditions are mathematically different , although physically equivalent .
thus different textures may occur for the same substrate under strong anchoring conditions .
the elastic energy per unit length along the @xmath2 axis has contributions associated to the disclination lines and the distortions close to the cusps of the form @xcite : @xmath94 where @xmath95 is the total number of disclination lines in the nematic and the index @xmath96 runs over all the cusps on the substrate .
the next - to - leading contribution to the free energy per unit length is expected to be independent of the lengthscale @xmath5 .
thus , for large @xmath5 we expect only transitions between textures with the same leading - order contribution to the free energy given by eq .
( [ asymptotic - core ] ) . however , we will see that for moderate values of @xmath5 other transitions between textures may be observed .
for a nematic with a @xmath82-disclination line in the bulk , represented by the filled circle .
the vertical dotted line corresponds to the branch cut , where a discontinuity of @xmath83 in the orientational field is observed .
the dashed line corresponds to the boundary of @xmath97 . ]
analytical solutions of nematic textures in contact with patterned substrates are not available , in general .
there are exceptions , which may be obtained , for example , by using conformal mapping techniques @xcite .
otherwise , we have to resort to numerical methods .
however , the presence of disclination lines and/or singularities associated to the surface cusps requires special techniques . in this section
we will describe the method used to obtain the nematic textures in the presence of disclination lines and/or substrate cusps , which is based on the numerical methods used previously for the sawtooth substrate @xcite .
first , we will assume that the number and positions of the disclination lines in the nematic are known . in this case
, the elastic contribution will be a function of the number of disclination lines and their corresponding positions .
their equilibrium values can be obtained by standard minimization techniques , such as conjugated - gradient methods .
we split the orientation field @xmath34 in two terms : a singular contribution @xmath98 , associated to the disclination lines in the nematic and/or the disclination - like singular contributions due to the surface cusps , and @xmath99 which we require to be regular everywhere in the integration domain @xmath58 .
if disclination lines are present in the nematic , we modify the integration @xmath58 to @xmath97 , excluding the defect core and the branch cut , which we will consider perpendicular as shown in fig .
we choose @xmath98 periodic in @xmath4 , with period @xmath5 , satisfying the laplace equation in @xmath97 ( @xmath58 in the absence of disclination lines ) , which captures the singularities in the orientational field associated to the disclination lines and surface cusps .
a possible choice , based in previous studies for the sawtooth substrate , is @xcite : @xmath100
\nonumber\\ & + & \sum_j i_j \bigg[\arctan\left(\frac{\tanh\frac{q}{2}(y - y_j)}{\tan\frac{q}{2}(x - x_j)}\right)\nonumber\\ & + & \arctan\left(\frac{1}{\tan\frac{q}{2}(x - x_j)}\right)\bigg ] \nonumber\end{aligned}\ ] ] where @xmath101 , the first sum runs over the surface cusps at @xmath102 , while the second one is over the disclination lines at positions @xmath103 in the nematic phase .
note that the contribution associated to the surface cusps vanishes as @xmath21 .
by contrast , the nematic disclination term exhibits a piecewise linear behaviour as @xmath21 and its elastic contribution to the interfacial free energy diverges unless the total topological charge associated to the nematic disclination lines vanishes .
therefore , we restrict our study to situations where the number of @xmath104-disclination lines in the nematic phase is the same as the number of @xmath105-disclination lines to ensure that the far - field director field is undistorted .
for the non - singular part @xmath99 , we have to solve the laplace equation subject to periodic boundary conditions on the sides @xmath106 and free boundary conditions as @xmath21 .
finally , @xmath99 satisfies dirichlet boundary conditions @xmath107 at the substrate .
as mentioned above , the regularity of @xmath99 in the region @xmath58 implies that standard numerical techniques can be used . in ref .
@xcite we used a boundary - element method in which the whole boundary of @xmath58 is discretized . in this paper
we use a different boundary - element method which only requires the discretization of the substrate relief .
the field @xmath99 can be written in @xmath58 as : @xmath108 g(\mathbf{s},\mathbf{r } ) \label{bem1}\\ & & -\theta_{ns}(\mathbf{s } ) \left[\boldsymbol{\nu}(\mathbf{s})\cdot \boldsymbol{\nabla}_{\mathbf{s } } g(\mathbf{s},\mathbf{r})\right ] \bigg ) \nonumber\end{aligned}\ ] ] where the contour integral over the boundary @xmath109 of region @xmath58 is counter - clockwise , @xmath43 is the outwards normal to the boundary at @xmath44 and @xmath110 is the fundamental solution of the laplace equation in the infinite strip @xmath111 , @xmath112 with periodic boundary conditions on @xmath4 : @xmath113 where @xmath114 and @xmath115 .
note that this solution can be obtained as the composition of the fundamental solution on the free plane @xmath116 and the conformal mapping @xmath117 which maps the strip onto the full complex plane .
as both @xmath99 and @xmath118 are periodic on @xmath4 with period @xmath5 , the contributions to the integral ( [ bem1 ] ) from the lateral sides @xmath119 cancel each other . on the other hand , @xmath120 at large @xmath9 . as we impose @xmath121 and @xmath122 as @xmath21 ,
the contribution to the integral ( [ bem1 ] ) due to the top boundary @xmath123 is equal to @xmath124 .
therefore , eq . ( [ bem1 ] )
can be rewritten as : @xmath125 g(\mathbf{s},\mathbf{r } ) \label{bem1 - 2}\\ & & -\theta_{ns}(\mathbf{s } ) \left[\boldsymbol{\nu}(\mathbf{s})\cdot \boldsymbol{\nabla}_{\mathbf{s } } g(\mathbf{s},\mathbf{r})\right]\bigg ) \nonumber\end{aligned}\ ] ] we impose dirichlet boundary conditions on the substrate relief , so the last term on the right - hand side of eq .
( [ bem1 ] ) is known .
the unknowns are the normal derivatives of @xmath99 on the substrate and the far - field value @xmath126 .
the former is obtained by solving the integral equation @xcite : @xmath127 g(\mathbf{s},\mathbf{s}_0)=\frac{\theta_{ns}(\mathbf{s}_0)-\theta_{ns}^\infty}{2 } \label{bem2 } \\ & & + \int_{\cal l } d \mathbf{s } \theta_{ns}(\mathbf{s } ) \left[\boldsymbol{\nu}(\mathbf{s})\cdot \boldsymbol{\nabla}_{\mathbf{s } } g(\mathbf{s},\mathbf{s}_0)\right ] \nonumber\end{aligned}\ ] ] where now @xmath128 . on the other hand , for large @xmath9 , @xmath129 and eq .
( [ bem1 - 2 ] ) reduces to @xmath130\left(\frac{s_y - y}{\lambda}+\frac{\ln 2}{2\pi}\right)\nonumber\\ & -&\frac{1}{\lambda}\int_{\cal
l } d \mathbf{s}\theta_{ns}(\mathbf{s})\nu_y(\mathbf{s } ) \nonumber\\ & = & \frac{1}{\lambda}\int_{\cal l } d \mathbf{s}\left ( s_y [ \boldsymbol{\nu}(\mathbf{s})\cdot \boldsymbol{\nabla}_{\mathbf{s } } \theta_{ns}(\mathbf{s } ) ] - \theta_{ns}(\mathbf{s})\nu_y(\mathbf{s})\right ) \label{bem3}\end{aligned}\ ] ] where @xmath131 and @xmath132 are the @xmath9-components of @xmath44 and @xmath51 , respectively , and the second equality comes from the fact that @xmath133 \label{condition_integral}\\ & = & \int_{\cal l } d \mathbf{s } [ \boldsymbol{\nu}(\mathbf{s})\cdot \boldsymbol{\nabla}_{\mathbf{s } } \theta_{ns}(\mathbf{s } ) ] \nonumber\end{aligned}\ ] ] finally , substituting eq .
( [ bem3 ] ) in eq .
( [ bem2 ] ) , we obtain : @xmath127 \left(g(\mathbf{s},\mathbf{s}_0)+\frac{s_y}{2\lambda}\right)=\frac{\theta_{ns}(\mathbf{s}_0)}{2 } \label{bem4 } \\ & & + \int_{\cal l } d \mathbf{s } \theta_{ns}(\mathbf{s } ) \left[\boldsymbol{\nu}(\mathbf{s})\cdot \boldsymbol{\nabla}_{\mathbf{s } } g(\mathbf{s},\mathbf{s}_0)+\frac{\nu_y}{2\lambda}\right ] \nonumber\end{aligned}\ ] ] in order to solve eq .
( [ bem4 ] ) , we discretize the boundary as a set of straight segments ( the boundary elements ) .
it is important to ensure that the substrate cusps correspond to extremes of these segments .
we use the constant boundary element approach @xcite , so we assume that both @xmath99 and its normal derivative are constant along each boundary element . introducing this approximation into eq . ( [ bem4 ] ) , we obtain a set of linear algebraic equations for the normal derivatives .
once this is solved , the far - field orientation @xmath126 can be obtained from eq .
( [ bem3 ] ) . the elastic contribution to the free energy per @xmath134unit length and @xmath135period
can be obtained from eq .
( [ frank3 ] ) as : @xmath136 due to the periodicity of @xmath8 on the boundaries @xmath106 and the free boundary at @xmath21 , the contour integral on the right - hand side can be reduced to : @xmath137 where the first integral is on the surface relief , and the other terms are on the branch cuts @xmath138 starting at the disclination @xmath139 position @xmath103 with topological charge @xmath140 . finally , we use the fact that @xmath141 in the derivatives , and the elastic contribution becomes : @xmath142 the first two terms exhibit singularities associated to the cusps ( first integral ) and disclination cores ( second term ) which must be handled carefully by deforming the contour with arcs of circle of radii @xmath69 to avoid them . in fact , these singularities lead to the contribution eq .
( [ asymptotic - core ] ) mentioned above . from a numerical perspective , the first integral can be obtained by considering the substrate relief discretization used to obtain @xmath99 .
thus , if we consider the boundary as the union of @xmath143 segments ( @xmath144 , ordered counterclockwise ) , the first integral may be approximated as : @xmath145 where @xmath146 and @xmath147 are the segment midpoint and the outwards unit normal to that segment @xmath143 , respectively .
substitution of the expression for @xmath98 eq . ( [ thetas ] ) into eq .
( [ energy4 ] ) leads after some algebra to : @xmath148 the first term in eq .
( [ energy5 ] ) corresponds to the contribution associated to the surface cusps . in the second term ,
the sum on @xmath96 is over the number of cusps , while the sum on @xmath149 is over the boundary elements . in this expression
, @xmath150 is the position of the cusp , @xmath151 is the effective topological charge of the cusp singularity , @xmath152 are the coordinates of the left extreme of the segment @xmath143 and @xmath153 with @xmath154 .
the prime denotes that we exclude from the sum the nodes which correspond to surface cusps .
finally , in the last term the sum on @xmath139 is over the disclination lines at positions @xmath103 and topological charge @xmath140 , while other terms have the same meaning as before . the second term in eq . ( [ energy3 ] )
can be obtained analytically as : @xmath155 where the sum on @xmath96 is over the surface cusps , the sums on @xmath139 and @xmath156 are over the disclination lines , and in the last term the prime denotes that we exclude from the sum terms with @xmath157 .
the last two contributions to eq .
( [ energy3 ] ) associated with @xmath99 can be obtained in a similar way .
the third term yields @xmath158 where @xmath159 and @xmath146 are the length and the value of @xmath8 at the midpoint of @xmath143 , respectively . on the other hand , @xmath160 is the value of the non - singular normal derivative obtained from eq .
( [ bem4 ] ) . finally , the last term in eq .
( [ energy3 ] ) can be obtained by standard integration techniques , where @xmath99 in the branch cut is evaluated using eq .
( [ bem1 - 2 ] ) .
the final ingredients are the core contributions @xmath84 arising from the disclination lines and the effective disclination - like singularities of the nematic textures .
these terms are not described by the macroscopic elastic theory , and we have to resort to a more microscopic description to evaluate them .
we will evaluate them using the mesoscopic landau - de gennes framework @xmath161 , where the bulk and elastic free energy densities are , respectively , @xmath162 the bulk term @xmath163 determines the bulk nematic order parameter : @xmath164 ( isotropic phase ) if @xmath165 , and @xmath166 ( nematic phase ) if @xmath167 .
the elastic term @xmath56 penalizes distorsions of the orientational field , with two elastic constants @xmath168 and @xmath169 related to the frank - oseen elastic constants : @xmath170 and @xmath171 . in addition
, we consider the surface free energy density used in refs .
@xcite : @xmath172 where @xmath173 is a parameter related to the anchoring strength @xcite and @xmath174 is the reference tensor order parameter on the substrate with cartesian components @xmath175 , and @xmath176 the cartesian components of the unit vector normal to the substrate @xmath51 .
we obtain the core contributions following the procedure described in ref .
note that , as the typical size of the cores is @xmath69 , the cores associated with the disclination lines in the nematic are independent from the substrate .
this is not the case at the cusps where we assume that two locally planar surfaces meet .
the core contribution of the cusp singularities depends , in general , on the anchoring strength . in what follows ,
we take the nematic to be at nematic - isotropic coexistence ( i.e. @xmath177 ) .
associated to the cusps of one substrate relief period are highlighted . ] in this section we will describe the results for different substrate reliefs . in particular , we will revisit the sawtooth case , and we will present results for the crenellated and the sinusoidal substrates . the sawtooth substrate has been studied for a number of years @xcite . we revisit some of the results reported previously .
we consider a symmetric sawtooth , characterized by a tilt angle @xmath178 and a side length @xmath179 , as shown in fig .
[ fig4](a ) , so that @xmath180 .
no bulk disclinations are expected in the nematic texture , at least at distances of order @xmath179 from the substrate .
it is found that there are two nematic textures that are locally stable : the @xmath181 texture , where the nematic field is oriented along the @xmath9 axis away from the substrate , in the far field , and the @xmath182 texture , where the nematic field is oriented along the @xmath4 axis in the far field .
these textures are characterized by different effective topological charges @xmath183 and @xmath184 associated to the top and bottom cusps , respectively . in the @xmath181 texture , @xmath185 and @xmath186 , while in the @xmath182 texture , @xmath187 and @xmath188 .
the elastic contribution to the interfacial free energy of the nematic ( per @xmath134 unit length and @xmath135period ) at this substrate was obtained analytically @xcite : @xmath189\gamma\left[\frac{1}{2}+\frac{\alpha}{\pi}\right]\right)\bigg ] \nonumber\end{aligned}\ ] ] with @xmath190 or @xmath191 , where @xmath192 numerical results are in excellent agreement with the analytical result ( see ref.@xcite for a more detailed discussion ) .
as @xmath193 , there is an anchoring transition from homeotropic to planar anchoring at @xmath194 .
however , the value of @xmath178 at the transition may be altered in two ways .
first , the core contributions @xmath84 associated to the effective disclination - like singularities may be different for the @xmath181 and @xmath182 textures . these core contributions will shift the transition ( from @xmath194 ) by a small amount , since @xmath195 for large @xmath5 . in the previous discussion we assumed that the splay and bend elastic constants are equal , in line with the observation that @xmath196 is very small for liquid crystals like 5cb , close to the nematic - isotropic phase transition . in order to estimate the effect of the elastic anisotropy on the anchoring transition , a perturbation theory around the one - elastic constant model was developed @xcite .
the first order correction in @xmath196 is obtained using eq .
( [ frank2 ] ) @xcite : @xmath197 where @xmath198 is the nematic orientation field for @xmath199 and @xmath200 .
the main contribution arises from the neighbourhood of the cusps , which leads to an additional term proportional to @xmath201 .
we refrain from giving explicit expressions for this leading - order correction , which implies that the anchoring transition shifts to @xmath202 if @xmath203 , and below @xmath204 otherwise .
physically this is due to the fact that the elastic distortions in the @xmath181 texture are mainly splay , while bend dominates in the @xmath182 texture ( see fig . 3 in ref.@xcite ) .
as a function of @xmath178 for different values of @xmath67 . ] finally , @xmath181 and @xmath182 are not the only textures that are possible at a sawtooth substrate . as mentioned in section [ sec2 ] , the effective topological charges associated with the disclination - like singularities arising from the substrate cusps may be expressed as @xmath205 , where @xmath206 is one possible value of the topological charge and @xmath67 is an integer .
therefore , there is an infinite number of ( pairs ) of topological charges @xmath184 and @xmath183 , since @xmath207 and @xmath208 and the periodicity requirement on @xmath8 requires @xmath209
. the boundary conditions on @xmath8 are then @xmath210 on the left - to - right uphill segments , and @xmath211 on the downhill segments .
the far - field value @xmath212 is the average of these , @xmath213 , which is the value of @xmath8 along the vertical lines emerging from the substrate cusps .
note that if @xmath214 we obtain the @xmath182 texture , while if @xmath215 the anchoring is oblique .
the latter may be solved using the schwarz - christoffel conformal mapping used for the @xmath181 and @xmath182 cases , leading to an elastic contribution to the interfacial free energy given by eq .
( [ theoreticalf ] ) , with @xmath216 defined as : @xmath217 \label{defkalpha2}\\ = \frac{k\pi}{\left(\frac{\pi}{2}\right)^2-\alpha^2}\left(m\frac{\pi}{2}-\alpha\right)^2= \mathcal{k}_0(\alpha ) \left(\frac{m\pi}{2\alpha}-1\right)^2 \nonumber\end{aligned}\ ] ] fig .
[ fig5 ] illustrates @xmath216 as a function of @xmath178 .
the lowest curves , with @xmath190 and @xmath214 , correspond to the @xmath181 and @xmath182 textures , respectively .
the other curves describe higher elastic energy states , and may be discarded at equilibrium .
a similar behaviour was found for isolated wedges @xcite . [
cols="^,^ " , ] we now turn to a sinusoidal substrate of period @xmath5 and amplitude @xmath218 . this has also been studied previously @xcite . as the substrate relief
does not have cusps , a state without defects is expected to be the least free - energy state .
this state exhibits homeotropic anchoring , i.e. the far - field nematic director is oriented along the @xmath9 axis , for all @xmath219 ( see fig .
[ fig10](a ) ) .
we denote this texture by @xmath181 . as @xmath219 increases , the substrate roughness increases , and the orientational field exhibits large distorsions to follow the anchoring at the substrate .
numerical results show that , under these circumstances , the elastic distortions are lowered by reorienting the nematic director field , in the groove , along the @xmath4 axis ( see fig .
[ fig10](b ) ) , and thus the texture exhibits planar anchoring , denoted by @xmath182 .
this texture involves the nucleation of two disclination lines with opposite topological charges , located by symmetry above the top and bottom of the substrate relief , at a distance plotted in fig .
[ fig11 ] .
this distance is proportional to @xmath5 , and in the limit of large @xmath5 , the disclination lines are not affected by the substrate .
the distance decreases as the substrate roughness increases until the it stabilizes for @xmath220 .
thus , in an effective way , the disclinations lines are bound to the surface relief ( on the @xmath5 scale ) , driving the orientational field almost horizontal everywhere . and @xmath221 , respectively , as a function of @xmath219 . ] ( continuous line ) and @xmath182 textures with @xmath222 ( dashed line ) , @xmath223 ( dotted line ) and @xmath224 ( dot - dashed line ) . ] at large @xmath5 , the interfacial free energy of the @xmath181 texture depends on @xmath218 and @xmath5 through the factor @xmath219 which determines the substrate roughness .
on the other hand , from eq .
( [ asymptotic - core ] ) the interfacial free energy of the @xmath182 texture has a leading contribution @xmath225 , and the next - to - leading term has the same @xmath219 dependence as above .
note that , in this case , we have to add the core contributions associated to the @xmath82 disclinations lines , with constant values @xmath226 and @xmath227 .
thus , at large @xmath5 only the @xmath181 texture is expected for any substrate roughness .
however , the weak dependence of the interfacial free energy of the @xmath182 texture on @xmath5 implies that , for moderate values of @xmath5 , an anchoring transition between the @xmath181 and @xmath182 textures may be observed .
[ fig12 ] shows the interfacial free energy of the @xmath181 and @xmath182 textures as a function of @xmath219 , for different values of @xmath5 . while the @xmath181 branch depends only on @xmath219 and
is an increasing function of this parameter , the @xmath182 branches are decreasing functions of @xmath219 , and for different values of @xmath5 are shifted by the @xmath228 term . fig .
[ fig13 ] shows the @xmath229 anchoring phase diagram .
the homeotropic anchoring state @xmath181 is favoured at low @xmath219 , while at large substrate roughness planar anchoring is observed , i.e. the @xmath182 has the lowest free energy .
we note that the value of @xmath5 at the transition increases almost exponentially with @xmath219 .
finally , as in the previous cases we can include the effect of the anisotropy of the elastic constants perturbatively .
if we assume that the elastic distortions are on the @xmath38 plane , the conclusion is that the anchoring transition , which corresponds to moderate values of @xmath5 , may be shifted by this contribution , although qualitatively it will be very similar .
however , if the twist elastic constant is smaller than @xmath230 , there is experimental @xcite and numerical @xcite evidence of a twist instability which breaks the azimuthal symmetry : the disclination line is no longer parallel to the @xmath2 axis , but exhibits a zig - zag structure which decreases the splay and bending distortions
. this can not happen in the sawtooth and crenellated substrates , since the disclination - like singularities are located at the surface cusps . and the substrate period @xmath5 . ]
in this paper we report the results of a numerical investigation of the equilibrium nematic textures at patterned substrates under strong anchoring conditions .
we used a modified frank - oseen model that accounts for the presence of disclination - like singularities near surface cusps , as well as disclination lines in the nematic .
we considered three substrate reliefs : the sawtooth substrate , the crenellated substrate and the sinusoidal subtrate .
for the sawtooth and sinusoidal substrates , we observed a homeotropic to planar anchoring transition as the substrate roughness is increased . on the other hand ,
the crenellated substrate exhibits a more complex anchoring phase diagram , with a homeotropic to oblique anchoring transition , which depends not only on the substrate roughness but also on the surface anchoring strength .
the latter results from the dependence of the core contribution of the cusp singularities on the anchoring strength .
although we used the landau - de gennes model to obtain the defect core contributions to the free energy , any other model could be considered .
this may change the results quantitatively , when this contribution is relevant as for the crenellated substrates , but not qualitatively . secondly
, our procedure can be easily modified to consider the presence of nematic - isotropic interfaces .
this allows the study of wetting , filling and related interfacial phenomena for nematic liquid crystals .
this is ongoing work , and will be published elsewhere .
finally , we restricted the nematic orientational distortions to the plane perpendicular to the longitudinal axis of the surface . the generalization to full three - dimensional systems to consider situations where twist @xcite or saddle - splay @xcite distorsions play a role is a formidable task which is currently beyond the scope of our work .
we acknowledge financial support from the portuguese foundation for science and technology under contracts nos .
excl / fis - nan/0083/2012 and uid / fis/00618/2013 ( nms and mmtg ) .
j.m.r .- e . also acknowledge partial financial support from the spanish ministerio de economa y competitividad through grant no .
fis2012 - 32455 , and junta de andaluca through grant no .
p09-fqm-4938 , all co - funded by the eu feder .
is calculated .
right panel : mapping of the minimization cell under the schwarz - christoffel transformation . ] in this appendix we will evaluate the elastic free energy of the symmetric textures on a crenellated substrate characterized by a block of width @xmath231 and height @xmath232 , and period @xmath233 . we note that , due to the symmetry of the texture , we may evaluate the elastic free energy in the domain shown in fig .
[ fig_appendix](a ) , which corresponds to half a period of the substrate relief .
the domain boundary follows the substrate except close to its cusps , where it is rounded by arcs of circle of radii @xmath69 .
the vertical lateral sides run from the subtrate to infinity .
we set the origin on the lower substrate cusp , so the upper substrate cusp is at @xmath234 . on the domain boundary
we set @xmath235 except for the vertical segment which joins the two surface cusps , where we set either @xmath236 or @xmath237 ( depending on the symmetric texture considered ) .
we map the domain in the @xmath2-plane to the upper half @xmath238-plane by using the following schwarz - christoffel transformation : @xmath239 \nonumber\\ \textrm{with } \gamma=\arcsin\left[\textrm{sign}\left(\frac{\zeta}{\zeta - b}\right ) \sqrt{\frac{(a+b)\zeta}{a(\zeta - b)}}\right ] \nonumber\end{aligned}\ ] ] where @xmath240 and @xmath241 are complex constants , @xmath242 and @xmath243 real numbers , and @xmath244 and @xmath245 are the incomplete elliptic integral of the first and third kind , respectively .
the points @xmath246 , @xmath247 , @xmath248 and @xmath249 in the @xmath134plane are mapped onto @xmath250 , @xmath251 , @xmath248 and @xmath252 , respectively .
these conditions fix the values of @xmath240 , @xmath241 , @xmath253 and @xmath252 .
in particular , as @xmath254 , we find that @xmath255 . on the other hand , the real part of eq .
( [ sctransform2 ] ) diverges as @xmath256 , but the imaginary parts have well defined values which differ by @xmath257 .
consequently , @xmath258 .
finally , the values of @xmath253 and @xmath252 are determined by the requirement that the image of the point @xmath246 is @xmath250 in the @xmath238-plane .
this leads to the conditions : @xmath259 \nonumber\\ \frac{l_1}{l_1+l_2}=\frac{1}{\pi\sqrt{a+b}}\im\bigg[(1+b)k\left(\frac{a(1+b)}{a+b}\right ) \label{sctransform4}\\ + b\pi\left(\frac{a}{a+b}\bigg|\frac{a(1+b)}{a+b}\right)\bigg ] \nonumber\end{aligned}\ ] ] where @xmath260 and @xmath261 are the real and imaginary parts of @xmath2 , respectively , and @xmath262 and @xmath263 are the complete elliptic integrals of the first and third kinds , respectively .
the values of @xmath253 and @xmath252 must be obtained numerically from these conditions for given values of @xmath264 and @xmath265 . in order to obtain the orientational field @xmath8 ,
we solve the laplace equation in the upper half @xmath238-plane , with the real axis as boundary , except close to @xmath251 and @xmath248 , where it is rounded by arcs of circle of radii @xmath266 and @xmath267 , respectively .
the boundary conditions are @xmath268 for @xmath269 and @xmath270 , and @xmath271 for @xmath272 .
the solution is : @xmath273 \label{solution}\ ] ] where @xmath274 .
the orientational field in the original domain is @xmath275 , where @xmath276 .
thus the free energy per @xmath134unit length and @xmath135period is @xmath277 we can use eq .
( [ sctransform ] ) to relate @xmath266 and @xmath267 to @xmath69 @xcite , leading to : @xmath278 therefore , the free energy eq .
( [ mfofreeenergy ] ) can be recast as : @xmath279 note that this expression depends only on the geometric characteristics of the surface relief , and not on the boundary condition which determines the symmetric nematic texture .
thus , both symmetric textures have exactly the same elastic contribution to the free energy .
this is consistent with the results for the single step solution @xcite , which would correspond to our case in the limit @xmath280 and @xmath281 .
199 lee b - w and clark n a 2001 _ science _ * 291 * 2576 kim j - h , yoneya m and yokoyama h 2002 _ nature _ * 420 * 19 .
ferjani s , choi y , pendery j , petschek r g and rosenblatt c 2010 _ phys .
lett . _ * 104 * 257801 .
brown c v , towler m j , hui v c and bryan - brown g p 2000 _ liq .
* 27 * 233 .
uche c , elston s j and parry - jones l a 2005 _ j. phys .
phys . _ * 38 * 2283 .
elston s j and parry - jones l a 2006 _ liq .
cryst . _ * 33 * 697 .
romero - enrique j m , pham c t and patrcio p 2010 _ phys .
e _ * 82 * 011707 rojas - gmez o a and romero - enrique j m 2012 _ phys .
e _ * 86 * 041706 raisch a and majumdar a 2014 _ epl _ * 107 * 16002 .
ledney m f , tarnavskyy o s , lesiuk a i and reshetnyak v y 2016 _ liq .
_ 10.1080/02678292.2016.1197973 .
bramble j p , evans s d , henderson j r , anquetil c , cleaver d j and smith n j 2007 _ liq .
cryst . _ * 34 * 1059 .
patrcio p , pham c t and romero - enrique j m 2008 _ eur .
j. e _ * 26 * 97 patrcio p , romero - enrique j m , silvestre n m , bernardino n r and telo da gama m m 2011 _ mol . phys .
_ * 109 * 1067 patrcio p , silvestre n m , pham c t and romero - enrique j m 2011 _ phys .
e _ * 84 * 021701 patrcio p , tasinkevych m and telo da gama m m 2002 _ eur .
j. e _ * 7 * 117 silvestre n m , eskandari z , patrcio p , romero - enrique j m and telo da gama m m 2012 _ phys
e _ * 86 * 011703 lewis a h , garlea i , alvarado j , dammone o j , howell p d , majumdar a , mulder b m , lettinga m p , koenderink g h and aarts d g a l 2014 _ soft matter _ * 10 * 7865 .
davidson a j and mottram n j 2012 _ eur .
* 23 * 99 oseen h 1933 _ j. chem .
faraday trans .
ii _ * 29 * 883 .
frank f c 1958 _ disc .
faraday soc .
_ * 25 * 19 .
rapini a and papoular m 1969 _ j. phys .
( paris ) colloq . _ * 30 * c4 - 54 .
brebbia c a and domnguez j 1992 _ boundary elements : an introductory course , 2nd ed . _
( southampton : computational mechanics publications ) .
katsikadelis j t 2002 _ boundary elements : theory and applications _
( amsterdam : elsevier ) .
silvestre n m , romero - enrique j m and telo da gama m m 2016 _ submitted to j. phys : condens . matter _
jeong j , kang l , davidson z s , collings p j , lubensky t c and yodh a g 2015 _ proc .
natl . acad .
usa _ * 112 * e1837 davidson z s , kang l , jeong j , still t , collings p j , lubensky t c and yodh a g 2015 _ phys .
e _ * 91 * 050501(r ) nayani k , chang r , fu j x , ellis p w , fernandez - nieves a , park j o and srinivasarao m 2015 _ nature commun .
_ * 6 * 8067 | in this paper we revisit the problem of a nematic liquid crystal in contact with patterned substrates .
the substrate is modelled as a periodic array of parallel infinite grooves of well - defined cross section sculpted on a chemically homogeneous substrate which favors local homeotropic anchoring of the nematic .
we consider three cases : a sawtooth , a crenellated and a sinusoidal substrate .
we analyse this problem within the modified frank - oseen formalism .
we argue that , for substrate periodicities much larger than the extrapolation length , the existence of different nematic textures with distinct far - field orientations , as well as the anchoring transitions between them , are associated with the presence of topological defects either on or close to the substrate . for the sawtooth and sinusoidal case , we observe a homeotropic to planar anchoring transition as the substrate roughness is increased . on the other hand , a homeotropic to oblique anchoring transition is observed for crenellated substrates . in this case , the anchoring phase diagram shows a complex dependence on the substrate roughness and substrate anchoring strength . |
it has long been known that galaxies in the dense cluster environments systematically differ from those in the field in their morphology , stellar populations , gas fractions and gas distributions .
more interestingly , observations of galaxy clusters at intermediate redshift show that these properties change with redshift .
this could indicate that the dense cluster environment speeds up the evolution of galaxies .
various mechanisms affecting the evolution of galaxies have been suggested such as ram - pressure stripping , merging , tidal interaction , harassment or starvation . in spite of the abundance of statistical studies on clusters at intermediate redshifts ( e.g. * ? ? ?
* ; * ? ? ?
* ) there is a lack of indepth studies of individual galaxies which will further constrain the environmental effects in clusters . our goal is to do a detailed study of galaxies that are currently being affected by the cluster environment .
virgo is ideal for this purpose .
its nearness allows us to study details , and as a dynamically young cluster it shows a variety of processes at work to affect the galaxies .
different mechanisms can be traced by hi observations . since the first virgo survey in early 1980s
the sensitivity of the vla ( very large array ) has been significantly improved by almost a factor of 10 with a comparable resolution . as several virgo galaxies observed with a higher sensitisity show ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) , the newer hi data will provide much more detailed ( and possibly never seen ) structures .
we selected 41 spirals ( s0/a@xmath0sm ) in the virgo cluster ( figure 1 ) .
these galaxies cover a factor of 50 in mass of the cluster and span a wide range in star formation properties @xcite .
they are located throughout the cluster , from the dense region close to the center to the low density outer parts .
compared to previous surveys ( e.g. * ? ? ?
* ; * ? ? ?
* ) with brightness cutoff of @xmath1 in @xmath2 , our sample also contains fainter systems of @xmath3 mag which are more likely to be affected by stripping or gravitational interactions due to their low masses .
the observations of the 13 galaxies were made between february and may 2004 in vla c - short array .
we integrated @xmath48 hours on each source with a total bandwidth of 3.25 mhz and a channel separation of 24.4 khz ( 5.2 km s@xmath5 at 21 cm ) .
online hanning smoothing has been applied and the resulting velocity resolution is 10.4 km s@xmath5 .
we typically reached @xmath60.3 mjy per beam ( typically @xmath7 ) per channel ( 10.4 km s@xmath5 ) .
this corresponds to a 3@xmath8 surface density sensitivity of 4@xmath910@xmath10 @xmath11 . in some cases ( e.g. ngc 4294 , ngc 4299 , ngc 4383 and ngc 4694 )
the cubes were spatially convolved in order to bring out faint structures .
in figure 2 we present the total hi maps of 13 galaxies .
ngc 4351 , ngc 4396 , and ngc 4189 decline more sharply in hi surface brightness on the side toward the cluster center .
these galaxies are at intermediate distances from m87 ( [email protected] deg ) and are likely to be experiencing on - going icm - ism pressure .
extended tails are seen in a number of galaxies .
ngc 4294/9 , ngc 4351 , ngc 4396 , ngc 4424 , ngc4651 , and ngc4698 are found at a range of distances ( @xmath12 deg from m87 ) .
no obvious companions are found around these galaxies and none of their tails except possibly ngc 4424 and ngc 4651 seems to be related to tidal interactions .
especially we note that ngc 4294 and ngc 4299 which are only 0.1 deg apart from each other and at similar redshifts , show tails in the same direction , unlike what happens in tidal interactions .
even though we do not have a coherent three dimensional picture of virgo , it is worth mentioning that all the tails except for one case ( ngc 4651 ) are pointing away from the cluster center .
the southern extension in hi of ngc 4424 could be related to the giant elliptical ngc 4472 which is located 1.56 deg away to south , either through a tidal interaction or an icm - ism interaction @xcite .
ngc 4651 is a peculiar case in a sense that its optical tail and the gas tail are extended in opposite directions , to east for the stellar extension and to west for the gas tail , suggesting a minor merger .
gas accretion or tidal interaction is also found in some galaxies such as ngc 4383 ( a small gas blob in se ) and ngc 4694 ( tidally interacting with a low surface brightness system vcc 2062 ) .
there are several galaxies with truncated hi disks such as ngc 4064 , ngc 4424 , ngc 4569 , and ngc 4580 .
ngc 4424 and ngc 4569 are located in high density regions in a sub - cluster or in the center of the cluster ( ngc 4569 is discussed further in the following section ) and icm - ism pressure likely has caused the truncation in the gas .
however ngc 4064 and ngc 4580 are exceptional in a sense that both of them are located in low density environments with projected distances of 8.8 and 7.2 degrees from m87 ( 3.1 and 2.5 virial radii ; * ? ? ?
* ) , respectively .
the most likely explanation is that they have gone through the center and are on their way out .
even if galaxies have not gone through the the highest density regions , icm - ism interactions could still happen at further distances from the cluster center when the galaxy interacts with locally enhanced icm due to sub cluster - cluster merging as suggested by @xcite for ngc 4522 .
deep optical and h@xmath13 images have already been taken .
more recently we have been granted galex ( galaxy evolution explorer ) time for the entire sample of 41 galaxies .
the uv data will allow us to trace the timescales on processes at work .
combined by multiwavelength data and also compared with simulations eventually we will get more clear understandings of galaxy evolutions in the cluster environments .
ngc 4569 is one of the closest galaxies to the cluster center in our sample and it is severely deficient in hi ( @xmath14% of the normal hi for its size and type ) .
both the h@xmath13 and hi are truncated , and only extend to 30% of its stellar disk . an anomalous arm is seen to the west both in hi and h@xmath13 ( figure 3 ) and the hi arm seems to be interrupted at the location of the h@xmath13 starburst outflow nebulosity .
the hi arm extends for 3 arcmin(@xmath413 kpc ) while the h@xmath13 nebulosity near the minor axis on the nw side extends up to 6 kpc from the nucleus .
some h@xmath13 filaments are also observed on the se side , but only within @xmath15 of the nucleus , and there is no large - scale h@xmath13 nebulosity in the se like that in the nw .
these diffuse structures seen along the minor axis must be a product of a nuclear starburst and have been disrupted by icm pressure ( nebulosity as the superbubble ) .
the icm pressure must be stronger on the se to see that the nebulosity is weaker on this side and the more extended hi structure on the other side seems to be consistent with this picture .
this arm thus must be extra - planar , because it appears to be interacting with the minor axis starburst outflow , and because it lacks significant dust extinction , suggesting that it lies behind the disk .
the arm resembles the features that are seen in some phases of icm - ism interaction simulations @xcite it has been suggested that the stripped gas can form one extra - planar arm by the combination of wind pressure and galaxy rotation ( see also * ? ? ?
ngc 4396 also show signatures of extra - planar gas .
the kinematics of this gas are especially interesting ( figure 4 ) .
note how the velocities in the component north of the disk are all redshifted w.r.t .
nearby disk on both sides of minor axis suggesting non - circular motions , possibly indication of the presence of extra - planar gas .
simulations , estimating icm pressure in this region or studying stellar population will be helpful for sorting out the interaction parameters and history .
some other cases of extra - planar gas in virgo cluster galaxies have been studied in highly inclined systems .
there are several pieces of evidence that ngc 4522 is experiencing ongoing icm - ism stripping despite its location far ( but similar to the distance of ngc 4396 ) from the cluster center @xcite .
see also the case of ngc 4402 by crowl et al .
( 2004 ) .
biller , b. a. , jones , c. , forman , w. r. , kraft , r. , & ensslin , t. 2004 , astro - ph/0406132 ( accepted to apj ) .
cayatte , v. , van gorkom , j. h. , balkowski , c. , & kotanyi , c. 1990 , aj , 100 , 604 crowl , h. , kenney , j. d. p. , van gorkom , j. , & vollmer , b. 2004 , extra - planar gas conference proceeding dressler , a. et al . 1999 ,
apjs , 122 , 51 kenney , j. d. p. , van gorkom , j. , & vollmer , b. 2004 , aj , 127 , 3361 kenney , j. d. p. , hameed , s. , chung , a. , van gorkom , j. , & vollmer , b. 2004 , in preparation koopmann , r. , & kenney , j. d. p. 2004 , apj , in press ( october 1 , 2004 ) , astro - ph/0406243 phookun , b. , vogel , s. n. , & mundy , l. g. 1993 , apj , 418 , 113 phookun , b. , & mundy , l. g. 1995 , apj , 453 , 154 poggianti , b. m. et al .
1999 , apj , 518 , 576 schulz , s. , & struck , c. 2001 , mnras , 328 , 185 tully , r. b. , & shaya , e. j. 1984 , apj , 281 , 31 vollmer , b. , cayatte , v. , boselli , a. , balkowski , c. , & duschl , w. j. 1999 , a&a , 349 , 411 vollmer , b. , cayatte , v. , balkowski , c. , & duschl , w. j. 2001 , aj , 561 , 708 vollmer , b. , balkowski , c. , cayatte , v. , van driel , w. , & huchtmeier , w. 2004 , a&a , 419 , 35 vollmer , b. , beck , r. , kenney , j. d. p. , & van gorkom , j. 2004 , aj 127 , 3375 warmels , r. 1988 , a&as , 72 , 19 | we present preliminary results of vla hi imaging of selected virgo cluster galaxies .
the goal is to study environmental effects on galaxy evolution .
our sample of 41 galaxies is spread throughout the cluster and spans a wide range in star formation properties . here
we present the total hi maps of 13 galaxies .
we find a number of galaxies with extended hi tails , almost all pointing away from the cluster center .
truncated hi disks are found close to the center but also in the outer region .
some galaxies near the cluster center show compression of the gas on one side .
multiwavelength data of ngc 4569 and kinematics on ngc 4396 indicate that some of the hi is extra - planar .
these preliminary results on the hi morphology already suggest that a variety of environmental effects such as icm - ism interactions , harassment , tidal interactions or mergers may be at work to affect the evolution of galaxies . |
in high - dimensional time series analysis , the need to define time - varying patterns of sparsity in model parameters has proven challenging .
dynamic latent thresholding , introduced in @xcite , provides a general approach that induces parsimony into time series model structures with potential to reduce effective parameter dimension and improve model interpretations as well as forecasting performance .
the utility of various classes of latent threshold models ( ltms ) has been demonstrated in recent applied studies in macroeconomics @xcite and financial forecasting and portfolio decisions @xcite .
the scope of the approach includes dynamic regressions , dynamic latent factor models , time - varying vector autoregressions , and dynamic graphical models of multivariate stochastic volatility , and also opens a path to new approaches to dynamic network modeling @xcite .
this paper adapts the latent thresholding approach to different classes of multivariate factor models with a one main interest in dynamic transfer response analysis .
our detailed case - study concerns time - varying lag / lead relationships among multiple time series in electroencephalographic ( eeg ) studies . here the latent threshold analysis of such models induces relevant , time - varying patterns of sparsity in otherwise time - varying factor loadings matrices , among other model features .
we evaluate and compare two different classes of models in the eeg study , and explore a number of posterior summaries in relation to this main interest .
time series factor modeling has been an area of growth for bayesian analysis in recent years .
two key themes are : ( i ) dynamic factor models , where latent factors are time series processes underlying patterns of relationships among multiple time series ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) ; and ( ii ) sparse factor models , where the bipartite graphs representing conditional dependencies of observed variables on factors are not completely connected ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) , increasingly applied in problems of classification and prediction . here
we combine dynamics with sparsity .
some of the practical relevance of models with time - varying factor loadings is evident in recent studies ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) . as the number of variables and factors increase
, so does the need to induce sparsity in loadings matrices to reflect the view that variables will typically be conditionally dependent on only a subset of factors . in a time series setting , however , the patterns of occurrence of zeros in otherwise time - varying factor loadings matrices may also be time - varying .
one factor may relate to one particular variable with a time - varying loading over a period of time , but be insignificant for that variable in other time periods . thus the need to develop models of time - varying sparsity of loadings matrices in dynamic factor models .
all vectors are column vectors .
we use @xmath0 , @xmath1 , @xmath2 , @xmath3 , @xmath4 , for the normal , uniform , beta , gamma , and wishart distributions , respectively .
succinct notation for ranges uses @xmath5 to denote @xmath6 when @xmath7 e.g. , @xmath8 denotes @xmath9 .
the indicator function is @xmath10 and @xmath11 is the diagonal matrix with diagonal elements in the argument and hence dimension implicit .
elements of any @xmath12vector time series @xmath13 are @xmath14 , @xmath15 and those of any @xmath16 matrix time series @xmath17 are @xmath18 @xmath19
in a general setting , the @xmath20vector time series @xmath21 , ( @xmath22 ) is modeled as @xmath23 where : * @xmath24 is a @xmath25vector of predictor variables known at time @xmath26 ; * @xmath27 is the @xmath28 matrix of regression coefficients at time @xmath26 ; * @xmath29 is the @xmath30 vector of latent factors , arising from some underlying latent factor process over time ; * @xmath31 is the @xmath32 matrix of factor loadings at time @xmath26 ; * @xmath33 is the residual term , assumed zero - mean normal with diagonal variance matrix @xmath34 of volatilities @xmath35 at time @xmath36 complete specification requires models for @xmath29 , @xmath27 , @xmath31 and @xmath35 over time .
typically , @xmath37 , and models are identified via constraints on @xmath31 , such as fixing @xmath31 to have zeros above a unit upper diagonal : @xmath38 and @xmath39 for @xmath40 in section [ sec : modelsmandm+ ] , there is interpretable structure to @xmath29 and alternative assumptions are natural . special cases and assumptions now follow .
* constant and sparse factor models : * much past work uses constant coefficients @xmath41 and loadings @xmath42 the pure factor model , with @xmath43 and @xmath44 typically assumes the factors @xmath29 are zero - mean and independent , yielding a linear factor representation of the conditional variance matrix of @xmath45 sparsity in @xmath46 then begins development of more parsimonious models for larger @xmath47 ( e.g. * ? ? ?
* favar models : * when @xmath24 concatenates past values @xmath48 @xmath49 to lag @xmath50 and @xmath51 are constant , the model is a factor - augmented vector autoregression ( favar ) .
variants based on differing models for @xmath29 are becoming of increasing interest in macroeconomics @xcite .
* factor stochastic volatility models : * traditional bayesian multivariate volatility models have @xmath52 @xmath44 and @xmath53 where @xmath54 model completion involves stochastic volatility model for the @xmath55 and @xmath56 based on either log - ar(1 ) models or bayesian discounting ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* time - varying regression and factor loadings models : * variants of models with time - varying @xmath57 are well - used ( e.g * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
typically , the elements @xmath58 are ar(1 ) processes . within this class ,
random walk models have flexibility to adapt to change over time , while stationary ar(1 ) models can have longer - term predictive value and interpretation @xcite .
* process models for factors : * models of factor processes @xmath29 typically involve either conditionally independent factors over time , with or without time - varying conditional variances , or stationary vector autoregressive ( var ) models .
we highlight example models that incorporate elements noted in section [ sec : dfms ] , while being customized to the eeg study : a response variable is hierarchically linked to current and lagged values of an underlying latent process of scientific interest .
a first latent factor model is discussed , then extended with a time - varying vector autoregressive component ; these two models are customized examples of time - varying favar processes .
a dynamic transfer response factor model ( dtrfm ) relates the outcome variables to a foundational , scalar latent process @xmath59 by specifying @xmath29 to be a vector of recent values of this underlying scalar process .
each outcome variable relates to potentially several recent and lagged values of @xmath59 through time - varying loadings coefficients ; at any instant in time , these coefficients define the transfer response of the variable to the history of the underlying process . as the loadings vary in time
, the form of this response then naturally varies .
the basic structure of the model is described here . in , set @xmath43 for all @xmath36 suppose also that @xmath60 for some @xmath61 where the scalar series @xmath59 is modeled as a time - varying autoregressive ( tvar ) process of order @xmath62 .
that is , @xmath63 where @xmath64 is the vector of ar coefficients at time @xmath36 conditional on the variance elements @xmath65 and @xmath66 , the @xmath33 , @xmath67 and @xmath68 sequences are assumedly independent over time and mutually independent .
equation ( [ eq : delta ] ) indicates that the @xmath69 coefficients follow a vector random walk over time , permitting time variation but not anticipating its direction or form .
coupled with we have the traditional specification of a bayesian tvar@xmath70 model for the latent @xmath59 process . for the @xmath71 scalar response variable , the above model implies @xmath72 showing the transfer of responses from past values of @xmath59 via the possibly quite widely time - varying loadings @xmath73 , the latter specific to series @xmath74 for each @xmath75 model identification is straightforward . from , it is clear that an identification problem exists with respect to the lag / lead structure , i.e. , the time origin for the latent @xmath59 process , as well as the scale of @xmath59 relative to the @xmath76 fixing elements of one row of @xmath31 to specified values obviates this .
here we do this on the first row : for one factor ( lag ) index @xmath77 we set @xmath78 and @xmath79 for @xmath80 this way , @xmath81 is a direct , unbiased measurement of @xmath82 , subject to residual noise , so that we have formal identification and a quantitative anchor for prior specification . beyond the need for priors for model hyper - parameters ,
we need structures for the error volatility processes @xmath83 in and the tvar innovations variance process @xmath84 in . for routine analysis that is not inherently focused on volatility prediction , standard bayesian variance discount learning models effective
random walks whose variability is controlled by a single discount factor are defaults .
specified to describe typically slowly , randomly changing variances , the inverse gamma / beta bayesian model has the ability to track time - varying variances over time , and to deliver full posterior samples from relevant conditional posteriors for volatility sequences in mcmc analyses .
we use variance discount models here , based on standard theory in , for example , @xcite and @xcite ; these are simply specified via two discount factor hyper - parameters : @xmath85 , for each of the set of observation volatilities , and @xmath86 for the tvar innovations volatility .
substantive interpretation is aided by investigating the more detailed structure that theoretically underlies the latent tvar process @xmath87 specifically , well - known ( and well - exploited ) time series decomposition theory ( e.g * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) shows that , given the model parameters , the @xmath59 series has the decomposition @xmath88 where the @xmath89 are simpler component time series processes and @xmath90 are non - negative integers such that @xmath91 the values of these integers and the nature of the component processes depend on the model parameters @xmath69 . typically , slow variation over time in these yields stable numbers @xmath90 and the resulting processes are computable directly from ( posterior samples or estimates of ) the @xmath59 and @xmath92 the component processes @xmath93 have the ( approximate ) forms of time - varying autoregressive moving averages
tvarma(2,1 ) processes exhibiting quasi - periodic behavior : each @xmath93 is a stochastic sine wave whose amplitude , phase and frequency varies in time ; the time variation in frequency is directly related to that in @xmath69 , while the amplitude and phase variation is inherent and driven by the levels of variation controlled by @xmath94 further , posterior inferences for the time - varying frequencies , amplitude and phase are directly available from posterior simulations that generate samples of the @xmath59 and @xmath69 at each time . in parallel , each @xmath95 is a tvar(1 ) process , with time variation in short - term autocorrelations driven by that in @xmath69 . as with the @xmath96 we have direct access to posterior inferences on the tvar(1 ) parameters of these component processes from simulations of the posterior for @xmath97 at each time .
this decomposition therefore gives inferences on underlying time - frequency and short - term dependency structures underlying @xmath59 and its dynamic behavior . from it
follows that @xmath98 where , for each @xmath99 in the ranges displayed , @xmath100 thus the transfer response pattern defined by the time - varying factor loadings translates the nature of the inherent , underlying components of the driving
@xmath59 process to each of the output / response variables .
the above shows that this class of models provides broad scope for capturing multiple time - varying patterns of component structure including several or many components with dynamically varying time - frequency characteristics via a single latent process filtered to construct the latent factor vector process @xmath29 in the general framework .
the flexibility of these models for increasingly high - dimensional response series @xmath101 is then further enhanced through the ability of models with series - specific and time - varying loadings @xmath73 to differentiate both instantaneous and time - varying patterns in the transfer responses .
a direct model extension adds back a non - zero dynamic regression term to provide an example of time - varying favar models .
that is , with the dynamic factor component as specified via model m ,
suppose @xmath101 now follows where @xmath102 the @xmath103 matrix @xmath27 contains time - varying autoregressive parameters and @xmath104 that is , @xmath101 is dynamically regressed on the immediate past value @xmath105 as well as the underlying components of a driving latent process through the dynamic transfer response mechanism : we denote this as a tv - var(1 ) component of the model .
this extension of model m allows for the transfer response effects of the fundamental , driving @xmath59 process to be overlaid with spill - over effects between individual response series from one time point to the next , modeled by a basic tv - var(1 ) component .
this can be regarded as a model extension to assess whether the empirical tv - var component is able to explain structure in the response data not adequately captured by the structure dynamic factor component .
for increasingly large @xmath106 the tv - var(1 ) model component alone ( i.e. , setting @xmath107 implies what can be quite flexible marginal processes for the individual @xmath108 in contrast , the dynamic transfer response factor component while also quite flexible represents structurally related processes .
there is thus opportunity to for evaluation of the latter in the extended model m+ .
as the dimension @xmath109 of response variables and the number @xmath110 of effective latent factors increases , it becomes increasingly untenable to entertain models in which all loadings in @xmath31 are non - zero .
further , depending on context , it is also scientifically reasonable to entertain models in which one or more variables may relate in a time - varying manner to a particular element of the latent factor vector for some periods of time , but that the relationships may be practically negligible at other epochs .
this is the concept of dynamic sparsity : a particular @xmath73 may be non - zero over multiple , disjoint time periods , and adequately modeled by a specified stochastic process model when non - zero , but effectively zero in terms of the effect of @xmath111 on @xmath112 in other periods .
the same idea applies to dynamic regression and/or autoregressive parameters in @xmath113 analysis that permits this will allow for adaptation over time to zero / non - zero periods as well as to inference on actual values when non - zero .
this includes extreme cases when a @xmath73 may be inferred as effectively zero ( or non - zero ) over the full time period of interest .
dynamic latent thresholding @xcite addresses this question of _ time - varying sparsity _ in some generality ; this approach is now developed in our context of dynamic transfer response factor models .
we anchor the development on basic ar(1 ) process models for the free elements of the dynamic factor loadings matrix @xmath114 recalling that the first row of elements is constrained to fixed ( 0/1 ) values as noted in section [ subsec : dtrfm ] . for @xmath115
the @xmath73 are modeled via what we denote by the lt - ar(1 ) processes defined as follows : @xmath116 where the latent process @xmath117 is ar(1 ) with @xmath118 and where @xmath119 the processes are assumed independent over @xmath120 the latent threshold structure allows each time - varying factor loading to be shrunk fully to zero when its absolute value falls below a threshold @xmath121 .
this way , a factor loads in explaining a response only when the corresponding @xmath117 is `` large enough '' .
inference on the latent @xmath117 processes and threshold parameters make this data - adaptive , neatly embodying and yielding data - informed time - varying sparsity / shrinkage and parameter reduction .
the same approach applies to the time - varying autoregressive parameters in the extension to model m+ .
that is , the effective model parameters @xmath122 are modeled as thresholded values of ar(1 ) processes @xmath123 in precisely the same way as for the @xmath76 details are left to the reader as they follow the development for @xmath31 with simple notational changes
. it will be evident that prior specification for threshold parameters @xmath121 are key in defining practical models .
we can do this by referencing the expected range of variation of the corresponding @xmath117 process . under the ar(1 ) process model detailed above
, @xmath117 has a stationary normal distribution with mean @xmath124 and variance @xmath125 given the hyper - parameters @xmath126 this allows us to compute the probability that @xmath117 exceeds the threshold i.e. , the probability of a practically significant coefficient across any range of possible thresholds .
@xcite follow this logic to specify informative , structured priors for @xmath121 that depend explicitly on @xmath127 we use this specification here ; in particular , take conditional uniform priors @xmath128 for some @xmath129 direct evaluation then yields marginal ( with respect to @xmath121 ) sparsity probabilities @xmath130 where @xmath131 is the standard normal cdf .
this is trivially evaluated . for large @xmath132 , this is also very well approximated by @xmath133 ( this is extremely accurate for @xmath132 as low as 2 and practically relevant values of @xmath132 exceeding that ) .
the sparsity probability strictly decreases in @xmath132 and decays to values of about 0.25 , 0.20 and 0.15 , respectively , at about @xmath134 and 5.3 , respectively .
this gives us assessment of what a particular choice of @xmath132 implies in terms of overall expected levels of sparsity _ a priori_. in our studies , we find strong robustness in posterior inferences to specified values of @xmath132 above 3 or so , and use that value as a default .
note also that there is flexibility to customize the prior to use different @xmath132 values for each threshold , to cater for contexts in which we aim to favor higher thresholds ( and hence higher probabilities of zero parameter process values ) for some @xmath135 than for others .
mcmc computations extend and modify those developed for time - varying autoregressions and multivariate stochastic volatility factor models in @xcite .
the overall mcmc integrates a series of steps that use standard simulation components from bayesian state space models ( e.g. , * ? ? ?
* ; * ? ? ?
* ) and from traditional ( static loadings ) latent factor models @xcite .
customization here involves modifications to resample the latent tvar factor process in our dynamic transfer responses factor context , and other elements including metropolis hastings steps as in @xcite for the latent threshold components .
the appendix accompanying this paper describes key details , and notes how the mcmc directly extends previously described strategies for dynamic latent threshold models .
electroencephalographic ( eeg ) traces are time series of electrical potential fluctuations at various scalp locations of a human subject , reflecting the complex dynamics of underlying neural communication .
analysis of multichannel eeg traces is key to understanding the impact of electroconvulsive therapy ( ect ) , one of the most effective treatments known for major depression with electrically induced seizures in patients @xcite .
the convulsive seizure activity drives multichannel eeg traces and statistical interest is to model such multivariate time series in order to reveal underlying characteristics and effects of ect .
various models have been studied to explore features of eeg time series ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
univariate tvar models are well - used and proven as models of individual eeg channels ( e.g. * ? ? ?
* ; * ? ? ?
* ) ; they can adequately represent what can be considerable changes in the patterns of evolution of time - frequency structure in such series , as well as differences and changes in relationships across the eeg channels .
such studies highlight the need for multivariate models of the time - varying commonalities across channels , with latent process structure reflecting the inherent , underlying mechanisms of neural communication .
our analysis adapts the earlier approach of @xcite .
that work was the first bayesian approach to multivariate time series analysis that incorporated the key scientific feature of a single , focal latent process driving the eeg signals across multiple channels .
the authors used a novel dynamic distributed lag approach that aimed to capture time - varying lag - lead structures across the eeg channels , introducing a customized model specific to that context .
though effective , that approach was very specific and empirical the authors developed dynamic regressions of @xmath136 of the channels on the _ observed signal _ of one selected channel , the latter chosen as an empirical proxy for the underlying latent driving process @xmath87 the developments of the current paper provide a general , flexible and in terms of the specific goals of the dynamic lag / lead study almost perfectly suited context that can be seen , in part , as an outgrowth from that prior work . here
the identification of dynamically adaptive lag / lead structure is driven by combining time variation in non - zero factor loadings with the latent threshold approach .
the study here explores @xmath137-channel eeg times series recorded in one seizure of one patient , as reported and earlier analyzed in @xcite and @xcite .
the eeg channels are @xmath137 electrodes located around and over the patient s scalp ; see figure [ fig : map ] .
the original data set has sampling rate 256hz over a period of 1 - 2 minutes ; following and to compare directly with @xcite , we analyze the series subsampled every sixth observation after removing about 2,000 observations from the beginning ( up to a higher amplitude portion of the seizure ) yielding @xmath138 observations .
representative graphs of data on two of the channels over selected epochs appear in figure [ fig : eegdata ] .
visual inspection of the substantial time - varying , quasi - periodic trajectories of the data indicates that signals on some eeg channels are obviously delayed with respect to other channels , and the apparent delays ( lags ) vary substantially through time .
this is perfectly consistent with the dynamic patterns of relationships of individual channels ( the @xmath139 with an underlying seizure process ( the latent @xmath59 ) captured by our model structure ( section [ subsec : dtrfmcomps ] ) ; the latent @xmath59 process represents a range of dynamic quasi - periodicities characterizing multiple brain wave components overlaid by , and modified by , the induced seizure , and the time - varying lag / lead relationships among channels are represented by channel - specific and time - varying factor loadings , some of which may be negligible for all time or for periods of time , and relevant elsewhere .
series are measurements of electrical potential fluctuations taken in parallel at each of these locations , defining the eeg channels ( international 10 - 20 eeg system).,width=192 ] + our analysis summaries are based on @xmath140 effective lags , i.e. , the model has a @xmath141dimensional latent factor vector @xmath142 and the first row of @xmath31 set to @xmath143 as the basis for model identification .
this precisely parallels the setup in the empirical model of @xcite .
as discussed in section [ subsec : dtrfm ] , some constraints of this form are needed on elements of @xmath31 to formally identify the single latent factor process model . there is no loss of generality nor any superfluous structure imposed on the model here ; we could choose any element of the first row of @xmath31 to insert the 1 , with different choices simply shifting the implied time origin of the @xmath59 process . under this structure ,
the first eeg channel @xmath81 loads only @xmath144 , while the other channels have loadings in the first ( last ) two columns of @xmath31 related to the leading ( lagged ) values of the @xmath59 process .
our analysis takes the so - called vertex channel cz as series @xmath145 .
see figure [ fig : map ] .
this parallels the use of the observed data on this specific channel as an empirical factor process in @xcite .
the other channels are ordered from the centre out .
one further modeling detail relates to a modification for a further , subtler soft identification question .
the model so far implies that @xmath146 so the conditional variation expected in channel 1 is the sum of time - varying contributions from the @xmath59 process plus @xmath147 as in all state - space models with multiple components contributing to variability in observed data , distinguishing and partitioning the contributions requires care in prior specification ; the picture is complicated here as time variation in @xmath148 competes with the intricate dynamics of the @xmath69 and time variation in @xmath84 .
a specification that controls this more usefully in the current latent factor models is to fix as constant the measurement error in series 1 , i.e. , set @xmath149 constant over time .
this ensures the interpretation of @xmath150 as pure measurement error ( there being no reason to expect time variation in pure measurement error variances , as opposed to the characteristics of the underlying factor processes and transfer response / loadings parameters ) .
we do this , maintaining the stochastic variance discount model for the other @xmath151 @xmath152 the latter combine pure measurement error and any other identified changes in residual volatility across these channels .
then , posterior inferences indicating substantial patterns of time variation in the latter then indicate the ability of the discount models to account for relative variability not captured by the underlying , identified latent factor process . the mcmc analysis of section [ sec : computation ] is trivially modified ; a traditional inverse gamma prior on @xmath153 leads to an inverse gamma complete conditional posterior .
the analyses summarized are based on model order @xmath154 for the latent @xmath59 process . while formal order selection approaches could be entertained ( e.g. , * ? ? ?
* ; * ? ? ?
* ) , our philosophy based on applied work with tvar and related models in many areas is to fit successively larger models and assess practical relevant of resulting inferences .
here we fit the dtrfm with model orders up to @xmath155 and for each analysis examine the posterior estimates of components @xmath156 as detailed in section [ subsec : dtrfmcomps ] . with successively higher values of model order @xmath157
we find robust estimation of @xmath158 quasi - periodic components with estimated frequencies varying over time in ranges consistent with known ranges of seizure and normal brain wave activity .
model order @xmath154 is needed to identify these three components , and they persist in models of higher order ; in addition to their substantive relevant , the estimated components are sustained over time and vary at practically relevant levels in terms of their contributions to each of the eeg series . however , moving beyond @xmath154 leads to increasing numbers of estimated components that are very ephemeral , of low amplitudes and higher inferred frequencies beyond the range of substantive relevance .
this signals over - fitting as the model caters to finer aspects of what is really noise in the data . and
then disregarding such estimated noise components is certainly acceptable , we prefer to cut - back to the model order @xmath154 that identifies the main component structures without these practically spurious elements . model specification is completed with priors for hyper - parameters .
we take @xmath159 , supporting a range of values for @xmath160 and with prior mean for @xmath160 near 4.5 .
seizure eeg data typically range over 300 - 600 units on the potential scale , with sample standard deviations over selected epochs varying from 40 - 100 or more .
hence an expectation of measurement error standard deviation around 4 - 5 is consistent with prior expectations that measurement error constitutes in the range of 4 - 12% of the signal uncertainty in the traces . for the stochastic discount variance models ,
we set values of the discount factors as @xmath161 ; this is based in part on examination of analyses with various values , and consistent with relatively modest levels of volatility in variance components .
priors for the hyper - parameters of the latent ar(1 ) parameter processes are as follows : @xmath162 , @xmath163 , and @xmath164 independently , for @xmath165 .
this anticipates persistence in non - thresholded latent factor loadings , while allowing for some of the loadings to exhibit notable patterns of change . finally , we take @xmath166 and set @xmath167 in the conditional uniform priors for thresholds .
summaries here come from @xmath168 mcmc draws after a burn - in period of @xmath169 .
computations were performed using custom code in ox @xcite .
figure [ fig : factor ] displays time trajectories of the posterior means of the factor process @xmath59 , and the volatility @xmath170 of its driving innovations . the figure displays similar trajectories for the time - varying characteristic frequency and modulus for each of the three identified quasi - periodic components in @xmath59 , the @xmath171 of section [ subsec : dtrfmcomps ] .
the @xmath171 component of lowest frequency has oscillations in the so - called seizure or slow wave band , considerably decaying toward the end of the seizure episode .
notably , the other two inferred components have characteristic frequencies and moduli that are rather stable over time though exhibit minor variation . .
, width=480 ] .
, width=480 ] the frequency trajectories of the three quasi - periodic components show that each lies in one of the expected neuropsychiatric categories : the so - called _ delta _ band ( roughly 0 - 4hz ) , _ theta _ band ( 4 - 8hz ) , and _ alpha _ band ( 8 - 13hz ) @xcite .
each component process @xmath171 is defined by the corresponding characteristic frequency , while being broad - band in the spectral domain with time - varying spectra that can be understood to peak at the characteristic frequencies themselves .
the lowest - frequency component stays in the _ delta _ range and gradually decreases over time ; its modulus is close to one ( solid line in figure [ fig : factor](iv ) ) , which indicates a considerably persistent component ; this so - called delta - slow wave dominates the factor process during the course of the seizure , while its frequency content slides towards lower values towards the end of the seizure episode .
the other two quasi - periodic components lie in the _ theta _ and _ alpha _ ranges ; their moduli and amplitudes are lower than those of the dominant component over the whole seizure course , and show only limited changes over time .
these reflect known frequency ranges of normal brain signaling , being dominated through much of the period by the strong seizure waveform .
the innovations volatility rises in the initial part of the seizure to drive increased amplitude fluctuations throughout the central section , and then decays in later stages corresponding to the control and dissipation of the brain seizure effects .
these features are consistent with expected structure in the seizure process , and with the broad results of @xcite .
figure [ fig : tvar ] provides the trajectories of the posterior means and 95% credible intervals for the tvar coefficients @xmath172 .
all are markedly time - varying .
the 95% credible intervals are slightly wider during the late time periods , which feeds through to increased uncertainties in features of the quasi - periodic components ; figures [ fig : factor](v ) and ( vi ) , displaying the posterior means and 95% credible intervals of the frequency and modulus for the lowest - frequency component respectively , showing somewhat increased uncertainties towards the end of the seizure . for channel f7 , index @xmath173 , indicating the inferred probabilities of lag - lead structure in transfer response to @xmath59 as they vary over time.,width=480 ] to generate some insights into the nature of dynamic sparsity under the latent threshold model , we select one channel f7 at @xmath173, and plot the corresponding trajectories of the estimated posterior probabilities @xmath174 over time ; i.e. , the probability of a non - zero loading of channel f7 on each of the values @xmath175 for @xmath176 see figure [ fig : s ] where we indicate the loadings on @xmath177 by lead(@xmath178 ) and ( @xmath179 ) respectively , that on @xmath144 by sync , and those on @xmath180 by lag(@xmath181 ) and ( @xmath182 ) respectively .
the annotation here refers to lead / lag relative to the vertex location cz that reads - out an unbiased estimate of @xmath144 .
so a non - zero loading of f7 on @xmath59 , for example , defines a 2-period lead of that channel relative to the vertex , whereas a non - zero loading on @xmath183 represents a 1-period lag relative to the vertex channel cz , and so forth . from the figure
, it is clearly inferred that there is strong synchrony between f7 and cz in their transfer responses to fluctuations in @xmath59 based on the sync trajectory .
also , f7 also has a reasonable probability of responding to the latent factor process @xmath59 1-period ahead of cz , and almost surely does not lag cz in the transfer response over most of the time period , nor lead by more than 1 period until perhaps later in the seizure episode .
the ability of latent thresholding to adaptively indicate existence of non - zero loadings during some periods and not others , while also operating as a global variable selection mechanism as well , is nicely exemplified here .
figure [ fig : st ] provides a visual display of posterior probabilities @xmath174 across all the channels @xmath184 , and drawn at selected snapshots in time , with images created by linearly interpolating between the estimates at the electrode locations .
note that the model says nothing about spatial relationships between channels .
the marked patterns of shrinkage in the latent threshold model analysis does nevertheless indicate strong spatial relationships , while the relationships also show marked patterns of change over time .
for example , loadings of lead(+2 ) are commonly and globally shrunk to zero from left frontal to right occipital sites . the lead(+2 )
loadings around right frontal and prefrontal areas exhibit evolving degree of shrinkage .
similar changes are found in the parietal and occipital regions of lag(-2 ) loadings .
meanwhile , almost no shrinkage is found in the synchronized loadings except for the channel t3 ( left temporal ) .
interpolating from values at the channels @xmath185 each row corresponds to the values at a selected time point , and the columns represent the indices @xmath186 in relation to the transfer responses to @xmath175 for @xmath176,width=480 ] with the lag - lead structure at selected time points .
the row - column layout of images corresponds to that in figure [ fig : st ] .
, width=480 ] figure [ fig : beta ] is a companion to figure [ fig : st ] that exhibits aspects of estimated factor loadings with the lag - lead structure at selected time points .
the images represent estimates @xmath187 where @xmath188 if @xmath189 and zero otherwise .
recall that the factor loading of vertex channel cz is fixed at 1 for the basis and 0 for lagged / leaded times .
the estimates show strong patterns of positive spatial dependencies with cz at the synchronized state ( zero lag / lead ) , with concurrent loadings on the @xmath59 process decaying towards the exterior regions .
the approximate centroid of the higher loadings region moves from front to back through the course of the seizure , consistent with what is understood to be the typical progression of seizure waveforms @xcite . in the third row of the figure ( @xmath190 ) , the highest loadings appears at and near channel pz , and the parietal region exhibits rising intensity .
another higher intensity is detected around the right temporal area in lead(+2 ) and in the channel c4 in lead(+1 ) .
this indicates dynamics of the driving latent process exhibited earlier in right temporal / central areas and followed in the occipital region ; this spatial asymmetry in estimated transfer response loadings again links to experimental expectations for the progression of seizure activity . in the last row of the figure ( @xmath191 , a late stage of the seizure ) , the major lead / lag loadings diminish while the synchronized loadings persist .
two animated figures , available as online supplementary material , provide more insight into the patterns of variation over time in factor loadings , the differences across channels , and the nature of the dynamic latent thresholding in particular .
the first animation http://www.stat.duke.edu/~mw/.downloads/nakajimawest2016eeg/animate-st.avi[(linked at the external site here ) ] shows a movie of patterns of @xmath174 interpolating from values at the channels @xmath185 this shows how these patterns evolve over time @xmath26 , providing a dynamic display from which the snapshots in figure [ fig : st ] are selected at four specific times .
the second animation http://www.stat.duke.edu/~mw/.downloads/nakajimawest2016eeg/animate-betat.avi[(linked at the external site here ) ] shows the corresponding movie for the interpolated estimates of factor loadings @xmath192 over all time ; the snapshots in figure [ fig : beta ] are selected at four specific times .
the animations clearly show and highlight regions of the brain surface where there is very low or negligible probability of lag or lead effects of the @xmath59 process , other regions where sustained effects are very evident and regions in which there is more uncertainty about potential effects , together with inferences on the quantified lag / lead effects in terms of the temporal evolution of the spatial patterns in estimated factor loadings .
figure [ fig : sigma ] plots @xmath193 i.e. , estimated standard deviations of the idiosyncratic shocks in each channel @xmath185 each graph is roughly located at the corresponding electrode placement .
recall that @xmath194 , the innovation standard deviation for the channel cz , is assumed time - invariant , representing measurement error only , as part of the model specification to define and identify the latent driving process @xmath59 .
the model allows for potential variations over time in standard deviations at other channels , with opportunity to identify variability in the data not already captured through the time - varying loadings and latent process structure . across all channels
@xmath185 for clarity in presentation , the y - scale has been omitted ; each standard deviation is graphed on the scale of @xmath195 for comparability across channels .
recall that the anchor channel cz has constant standard deviation representing pure measurement error around the latent process at that channel .
each graph is roughly located at the corresponding electrode placement and the @xmath196-axes represent the full time period @xmath197.,width=307 ] from figure [ fig : sigma ] , there do appear to be variations across channels and they show some local spatial dependence .
trajectories of the neighboring channels f4 and f8 are clearly similar , exhibiting a major hike in the middle of the seizure .
it is evident that some parietal and occipital sites ( t3 , p3 , o1 , o2 and p4 ) share a common trajectory , which marks a peak in an early stage of the seizure then gradually decrease towards the end of the seizure . as seen in figures [ fig : st ] and [ fig : beta ] , these sites also share some relationships in the latent threshold - induced shrinkage and loadings at lag(-1 ) .
further , the estimate shows similarities among the channels pz , c4 and t6 , whose patterns differ from those in the occipital region .
this suggests an intrinsic difference between the central sites ( pz , c4 and t6 ) and the occipital sites ( t3 , p3 , o1 , o2 and p4 ) , also suggested by figures [ fig : st ] and [ fig : beta ] . across all but the vertex channel @xmath198 at @xmath199 the idiosyncratic error terms @xmath200
represent a compound of measurement error and of additional patterns including local sub - activity of the seizure that is not explained by the latent factor process .
there are also experimental and physiological noise sources that are likely to induce local / spatial residual dependencies in the data not forming part of the main driving process @xmath201 including electrical recording / power line noise and scalp - electrode impedance characteristics ; these presumably also contribute to the time - variation patterns in the @xmath83 identified and their spatial dependencies .
model m+ has @xmath202 where @xmath27 is the @xmath103 matrix of lag-1 time - varying coefficients modeled using latent threshold ar(1 ) processes .
model m+ extends model m to potentially capture data structure not fully explained by the factor and residual component .
one interest is to more structurally explain the time variation in estimated residual volatilities @xmath83 exhibited in the analysis of the baseline model m. a contextual question is that of representing potential spill - over effects between eeg channels as the seizure waves cascade around the brain ; that is , local ( in terms of the neural network and perhaps in part , though not necessarily , physically spatially ) transmission of signals between subsets of channels that represent delayed responses to the latent @xmath59 process not already captured by the dynamic latent factor model form .
the matrix @xmath27 is expected to be sparse and modeled via latent threshold dynamic models , as earlier described . across all channels @xmath184 for the extended model m+ .
details as in as in figure [ fig : sigma ] for model m.,title="fig:",width=480 ] + across all channels @xmath184 for the extended model m+ .
details as in as in figure [ fig : sigma ] for model m.,title="fig:",width=480 ] .4 in across all channels @xmath184 for the extended model m+ .
details as in as in figure [ fig : sigma ] for model m.,width=307 ] figure [ fig : at ] plots the posterior means of the states @xmath203 and the posterior probabilities @xmath204 , where @xmath205 and @xmath206 is the latent threshold for each state @xmath207 .
the matrix @xmath27 is evidently sparse and exhibits considerable changes in the state and the posterior shrinkage probability among the selected time points .
figure [ fig : sigma - var ] shows the estimated standard deviations of the idiosyncratic shocks @xmath208 .
compared with figure [ fig : sigma ] , the trajectories of standard deviations are generally somewhat smoother over time ; some of the variation in the data not already captured by the @xmath209 is now absorbed by the @xmath210 . to explore some practical implications of the extended model m+ and compare with the baseline model m
, one aspect of interest is predicted behavior of the time series based on _ impulse response analysis _ relative to the underlying @xmath59 process . standing at a current , specified time @xmath211
this simply asks about the nature of expected development of the series @xmath212 over the next @xmath213 time points based on an assumed level of the impulse
@xmath214 to the driving innovations of the latent process . in applied work in economics ,
impulse responses are often primary vehicles for communicating model implications , comparing models , and feeding into decisions .
the use of latent thresholding in macroeconomic models has focused on this , in part , and clearly demonstrated the utility of dynamic latent thresholding in inducing more accurate predictions and , in particular , more statistically and substantively reliable impulse response analyses @xcite .
we do this here from three time points @xmath215 chosen in the early , middle and later sections of the eeg series ; this exhibits differences in the model projections / implications over time due to the dynamics , as well as differences between the two models in each of these periods .
computations are easily done by using the posterior mcmc samples to project forward in time ; predictive expectations are then computed as monte carlo averages .
the impulse value @xmath216 is taken as the average over @xmath217 of the estimated historical innovations standard deviations @xmath218 . figure [ fig : imp ] plots the impulse responses of the 19 eeg channels with @xmath219 and from each of the two models .
note that , for our comparison purposes here , we are interested in the _ forms _ of the impulse responses over the horizon specified , not their specific values .
we already know that the innovations variance @xmath84 shows marked changes over time and , in particular , decays to rather low values in the later stages of the seizure .
hence the shock size @xmath216 taken here is larger than relevant and realized innovations over the latter part of the time period , and the amplitudes of impulse responses should therefore not be regarded as pertinent .
the form of the projections are the focus .
\(i ) model m + eeg channels to a shock to the underlying factor process @xmath59 obtained from ( i ) model m , ( ii ) model m+ . the impulse response functions computed at three different time points throughout the seizure are shown ( columns ) . for each model ,
the impulse response projections are made from the time point indicated by column header up to 80 time periods ahead ( lower rows in ( i ) and ( ii ) ) ; the same responses are shown on a truncated time period up to only 30 time periods ahead ( upper rows in ( i ) and ( ii ) ) , for clarity.,title="fig:",width=364 ] + ( ii ) model m+ + eeg channels to a shock to the underlying factor process @xmath59 obtained from ( i ) model m , ( ii ) model m+ .
the impulse response functions computed at three different time points throughout the seizure are shown ( columns ) .
for each model , the impulse response projections are made from the time point indicated by column header up to 80 time periods ahead ( lower rows in ( i ) and ( ii ) ) ; the same responses are shown on a truncated time period up to only 30 time periods ahead ( upper rows in ( i ) and ( ii ) ) , for clarity.,title="fig:",width=364 ] \(i ) @xmath220 + and @xmath221 , respectively.,title="fig:",width=384 ] + ( ii ) @xmath222 + and @xmath221 , respectively.,title="fig:",width=384 ] patterns of the impulse response are clearly time - varying across the three exhibited time points ; variation is evident with respect to wave frequency , persistence / decay speed , and variation across the channels . in early periods of the seizure ,
the responses decay slowly with a high - frequency cyclical wave , while in later periods the decay is more rapid and the oscillations at lower frequency .
while there are , as we have discussed above , marked patterns of variation in lag / lead relationships across channels , there is the appearance of stronger synchronicity earlier in the seizure episode , and this deteriorates towards end of the seizure .
the responses from the tv - var extended model m+ model exhibit more variation across the channels than those from the model m. this is attributable to the induction of some spill - over effects of the shock . through the latent factor model component alone
, the shock @xmath223 has an impact on each of the channels through its immediate influence on @xmath224 and the consequent transfer response of these effects via @xmath225 and so forth . in the extended model
, additional feed - forward effects are passed through the channels via the tv - var component .
some additional insights into the nature of impulse responses can be gained from figure [ fig : impulses ] that shows images interpolating the 9 channel responses across the brain areas , based on analysis of the extended model m+ .
these are shown at six time points across the seizure period , and for selected horizons @xmath220 and @xmath221 , respectively ; these images clearly show the time variation of the responses spreading over the channels .
an animated figure , available as online supplementary material , provides a dynamic display over impulse response horizons 1:80 , with a movie that more vividly exhibits the differences due to time period .
the animations http://www.stat.duke.edu/~mw/.downloads/nakajimawest2016eeg/animate-impulse.avi[(linked at the external site here ) ] represent the six time points in figure [ fig : impulses ] , and show images of the impulse responses as the projections are made over @xmath226 to horizon @xmath227
the eeg time series analysis highlights the utility of latent thresholding dynamic models in constraining model parametrization adaptively in time , with resulting improvements in intepretation and inferences on inter - relationships among series and transfer response characteristics .
an additional comment on model comparison in the case study is worth mentioning .
statistical evaluation and comparison of model m@xmath228 with model m is implicit since the latter is a special case of the former .
the analysis results of m@xmath228 explicitly show the relevance of the extensions and hence support the more general model .
this is separately supported by values of the deviance information criterion ( dic ; see @xcite ) computed from the mcmc results for each model separately ; this yields estimated dic is 996,191.7 for model m and 988,435.9 for model m@xmath228 , which indicates strong evidence that model m@xmath228 dominates model m. a number of methodological and computational areas remain for further study . among them , we note potential for integrating spatially - dependent structures with latent threshold factor models , motivated in part by the spatial - temporal findings in the eeg study . also , incorporating two or more common latent processes might allow evaluation of more complex latent factor structures for these and other applications .
computational challenges are clear in connection with applying these models to higher dimensional time series such as are becoming increasing common in neuroscience as they are other other areas . that said , we expect the dynamic latent thresholding approach to become increasing relevant and important in constraining and reducing effective parameter dimension via dynamic sparsity in model parameters in contexts with higher - dimensional time series .
based on the observations @xmath8 , the full set of latent state parameters and model parameters for the posterior analysis of dtrfm model m is as follows : * the latent factor process states @xmath229 including uncertain initial values ; * the latent tvar coefficient process @xmath230 ; * the variance processes @xmath231 and @xmath65 ; * the latent factor loading process @xmath232 , including the uncertain initial state ; * the hyper - parameters @xmath233 and @xmath66 ; * the latent threshold hyper - parameters @xmath234 . the model of can be written in a conditionally linear , gaussian dynamic model form with a modified state @xmath235 and a state transition @xmath236 where @xmath237 generation of the full sets of states is obtained by the standard forward filtering , backward sampling ( ffbs ) algorithm ( e.g. * ? ? ?
* ) , which is efficient in the sense that the full trajectories of the states over time are regenerated at each iterate of the overall mcmc .
conditional on @xmath238 and the variances @xmath239 , reduce to a univariate , linear and gaussian dynamic regression model with respect to the state process @xmath240 .
we sample the states using the ffbs algorithm . based on the standard inverse gamma / beta bayesian discount model for the variance sequence @xmath65 over time
as noted in section [ subsec : dtrfm ] , the corresponding ffbs for volatilities provides a full sample from the conditional posterior for @xmath65 given all other quantities .
similarly , the full conditional posterior for the @xmath231 factorizes into @xmath109 components involving the individual @xmath241 separately over @xmath74 and the discount variance ffbs applies to each in parallel to generate full conditional posterior samples .
following @xcite , we sample each @xmath242 from its conditional posterior distribution given @xmath243 and all other parameters . recall that the elements of @xmath244 follow standard ar(1 ) processes , but are linked to the observation equation by the latent threshold structure .
the resulting conditional posterior for @xmath244 is a non - standard distribution that we can not directly sample .
we use a metropolis - within - gibbs sampling strategy with the proposal distribution derived in the _ non - threshold _
case by assuming @xmath245 ; i.e. , we generate the candidate from a standard linear dynamic model for @xmath244 without the latent thresholds ( see section 2.3 of * ? ? ?
priors for the latent ar hyper - parameters @xmath246 assume prior independence across series @xmath247 with traditional forms : normal or log - gamma priors for @xmath124 , truncated normal or shifted beta priors for @xmath248 and inverse gamma priors for @xmath249 . on this basis ,
the full conditional posterior for @xmath246 breaks down into conditionally independent components across @xmath250 we then resample the @xmath251 in parallel across @xmath74 using direct sampling from the conditional posterior in cases that the priors are conditionally conjugate , or alternatively via metropolis hastings steps .
as discussed in section [ subsec : dtrfm ] , the structured prior for the thresholds @xmath121 takes them as conditionally independent over @xmath115 with marginal priors that depend on the parameters of the corresponding latent ar processes , viz .
@xmath252 where @xmath125 the set of thresholds are then also independent in the complete conditional posterior ; they are resampled in parallel via metropolis hastings independence chain steps using the conditional uniform priors as proposals .
this is precisely as pioneered in @xcite in other latent threshold models , and its efficacy has been borne out in a number of examples there .
finally , note that the above requires a slight modification and extension to generalize the mcmc for the extended dtrfm model m+ of section [ subsec : tvvardtrfm ] .
the extension now involves the tv - var parameter matrices @xmath253 in with @xmath254 together with the required latent initial missing vector @xmath255 the above development applies conditional on these elements with the obvious modifications to subtract @xmath210 from @xmath101 throughout .
then additional mcmc steps are needed .
first , @xmath256 is generated from a complete conditional normal posterior under a suitably diffuse normal prior .
second , the latent thresholded elements of the sequence @xmath257 and the set of hyper - parameters of the underlying ar(1 ) processes as well as the corresponding thresholds , are treated just as are the elements of @xmath258 discussed above .
this component is a special case of the mcmc analysis for more general tv - var models as developed in @xcite .
some insights into the convergence of the mcmc sampling are gained by viewing trace plots for selected parameters . as an example
, some such plots from the analysis of the extended model m@xmath228 are shown in figure [ fig : trace ] . | we discuss bayesian analysis of multivariate time series with dynamic factor models that exploit time - adaptive sparsity in model parametrizations via the latent threshold approach .
one central focus is on the transfer responses of multiple interrelated series to underlying , dynamic latent factor processes .
structured priors on model hyper - parameters are key to the efficacy of dynamic latent thresholding , and mcmc - based computation enables model fitting and analysis . a detailed case study of electroencephalographic ( eeg ) data from experimental psychiatry highlights the use of latent threshold extensions of time - varying vector autoregressive and factor models .
this study explores a class of dynamic transfer response factor models , extending prior bayesian modeling of multiple eeg series and highlighting the practical utility of the latent thresholding concept in multivariate , non - stationary time series analysis .
_ msc 2010 subject classifications : _
62f15 , 62m10 , 62p10 _ key words & phrases : _ dynamic factor models ; dynamic sparsity ; eeg time series ; factor - augmented vector autoregression ; impulse response ; multivariate time series ; sparse time - varying loadings ; time - series decomposition ; transfer response factor models . |
stabilizer states are ubiquitous elements of quantum information theory , as a consequence both of their power and of their relative simplicity .
the fields of quantum error correction , measurement - based quantum computation , and entanglement classification all make substantial use of stabilizer states and their transformations under clifford operations @xcite .
stabilizer states are distinctly quantum mechanical in that they can possess arbitrary amounts of entanglement , but the existence of a compact description that can be updated efficiently sets them apart from other highly entangled states .
their prominence , as well as their name , derives from this description , a formalism in which a state is identified by a set of pauli operators generating the subgroup of the pauli group that stabilizes it , i.e. , the subgroup of which the state is the @xmath0 eigenvector . in this paper
we seek to augment the stabilizer formalism by developing a graphical representation both of the states themselves and of the transformations induced on them by clifford operations .
it is our hope that this representation will contribute to the understanding of this important class of states and to the ability to manipulate them efficiently .
the notion of representing states graphically is not new .
simple graphs are regularly used to represent _ graph states _ , i.e.
, states that can be constructed by applying a sequence of controlled-@xmath1 gates to qubits each initially prepared in the state @xmath2 .
the transformations of graph states under local clifford operations were studied by van den nest @xcite , who found that local complementation generated all graphs corresponding to graph states related by local clifford operations .
the results presented here constitute an extension of work by van den nest and others to arbitrary stabilizer states .
our graphical depiction of stabilizer states is motivated by the equivalence of stabilizer states to graph states under local clifford operations @xcite .
because of this equivalence , _ stabilizer - state graphs _ can be constructed by first drawing the graph for a locally equivalent graph state and then adding decorations , which correspond to local clifford operations , to the nodes of the graph .
only three kinds of decoration are needed since it is possible to convert an arbitrary stabilizer state to some graph state by applying one of six local clifford operations ( including no operation ) to each qubit .
the standard form of the generator matrix for stabilizer states plays a crucial role in the development of this material , particularly in exploring the properties of _ reduced graphs _ , a subset of stabilizer graphs ( which we introduce ) that is sufficient for representing any stabilizer state .
more generally , however , our stabilizer - state graphs are best understood in terms of a canonical circuit for creating the stabilizer state .
this description also permits the use of circuit identities in proving various useful equalities . in this way
, we establish a correspondence between clifford operations on stabilizer states and graph operations on the corresponding stabilizer - state graphs .
ultimately , these rules allow us to simplify testing the equivalence of two stabilizer graphs to the point that the test becomes provably trivial .
this paper is organized as follows .
section [ sec : background ] contains background information on stabilizer states , clifford operations , and quantum circuits .
stabilizer - state graphs are developed in sec .
[ sec : graphs ] , and a graphical description of the action of local clifford operations on these graphs is given in sec .
[ sec : transformations ] .
the issue of the uniqueness of stabilizer graphs is taken up in sec .
[ sec : equiv ] .
the appendix deals with the graph transformations associated with @xmath3 gates .
the _ pauli group _ on @xmath4 qubits , @xmath5 , is defined to be the group , under matrix multiplication , of all @xmath4-fold tensor products of the identity , @xmath6 , and the pauli matrices , @xmath7 , @xmath8 , and @xmath9 , including overall phases @xmath10 and @xmath11 . a _ stabilizer state _
is defined to be the simultaneous @xmath0 eigenstate of a set of @xmath4 commuting , hermitian pauli - group elements that are independent in the sense that none of them can be written as a product of the others .
these elements are called _ stabilizer generators _ and are denoted here by @xmath12 , while @xmath13 is used to denote the @xmath14th pauli matrix in the tensor - product decomposition of generator @xmath12 .
stabilizer generator sets are not unique ; replacing any generator with the product of itself and another generator yields an equivalent generating set .
an arbitrary product of stabilizer generators , @xmath15 , where @xmath16 is called a _ stabilizer element _ ; the stabilizer elements make up a subgroup of the pauli group known as the _
stabilizer_. a _ graph state _ is a special kind of stabilizer state whose generators can be written in terms of a simple graph as @xmath17 where @xmath18 denotes the set of neighbors of node @xmath19 in the graph ( see sec . [
subsec : terminology ] and ref .
@xcite for graph terminology ) .
_ simple graphs _ and , hence , graph states can also be defined in terms of an _ adjacency matrix _ @xmath20 , where @xmath21 if @xmath22 and @xmath23 otherwise . in a simple graph ,
a node is never its own neighbor , i.e. , there are no self - loops ; thus the diagonal elements of the adjacency matrix of a simple graph are all equal to zero .
the binary representation of the pauli group associates a two - dimensional binary vector @xmath24 with each pauli matrix @xmath25 , where @xmath26 , @xmath27 , @xmath28 , and @xmath29 .
this association is generalized to an arbitrary element @xmath30 , whose @xmath14th pauli matrix is @xmath31 , by letting @xmath32 be a @xmath33-dimensional vector whose @xmath14th and @xmath34th entries are the entries of @xmath35 , i.e. , @xmath36_k & [ r(p)]_{n+k } \end{array } \right)$ ] . the binary representation of a pauli - group element specifies the element up to the overall phase of @xmath37 or @xmath38 ; hermitian pauli - group elements are specified up to a sign .
the binary representation of the product of two pauli - group elements , @xmath39 , is the binary sum of their associated vectors , i.e. , @xmath40 .
two such elements commute if their _ skew product _ , @xmath41_j[r(q)]_{n+j}+[r(p)]_{n+j}[r(q)]_j\;,\ ] ] has value @xmath42 ; otherwise they anticommute . using binary notation , a full set of generators for a stabilizer state can be represented ( up to a sign for each generator ) by an @xmath43 _ generator matrix _ whose @xmath19th row is @xmath44 . because the stabilizer generators commute , the rows of the generator matrix are orthogonal under the skew product .
similarly , the independence of the stabilizer generators under matrix multiplication implies that the rows of the generator matrix are linearly independent under addition .
the freedom to take products of stabilizer generators without changing the stabilized state becomes , for the generator matrix , the freedom to add one row of the matrix to another .
the exchange of any pair of qubits @xmath19 and @xmath14 of a stabilizer state corresponds to the exchange of columns @xmath19 and @xmath14 and columns @xmath45 and @xmath46 in the generator matrix . since rows of the generator matrix are linearly independent and have vanishing skew product , these two operations are sufficient to allow us to transform any generator matrix to a _
canonical form _
@xcite , @xmath47 where @xmath48 and the vertical line divides the matrix in half .
the vanishing skew product of rows of the generator matrix implies that @xmath49 is a symmetric matrix and that @xmath50 and @xmath51 appear as indicated .
the @xmath6s in eq .
( [ eq : canonicalstabilizer ] ) denote a pair of identity matrices whose dimensions sum to @xmath4 .
the dimension of the upper left identity matrix is called the _ left rank _ of the generator matrix . due to the freedom inherent in qubit exchange ,
the canonical form of the generator matrix is not unique . for graph states , eq .
( [ eq : canonicalstabilizer ] ) becomes @xmath52 where @xmath49 has only @xmath42s on the diagonal .
graph states thus have generator matrices of full left rank , with @xmath49 being the adjacency matrix of the graph state s underlying graph .
we denote generator matrices of this sort by the term _ strict graph form _ , whereas the term _ graph form _
is used for generator matrices of the form shown in eq .
( [ eq : canonicalgraph ] ) where @xmath49 is any symmetric matrix .
the binary representation does not encode the sign of pauli group elements , so the generator matrix really specifies a set of generators up to @xmath53 possible sign assignments and thus specifies not a single stabilizer state , but rather an orthonormal basis of simultaneous eigenstates of the generators , each member of which corresponds to one of the sign choices . despite this , we continue , for convenience , to refer to _ the _ stabilizer state associated with a generator matrix .
the _ clifford group _ is the normalizer of the pauli group , i.e. , the set of all unitary operators @xmath54 such that @xmath55 for all @xmath56 .
any local operation in the clifford group can be obtained by repeated application of hadamard and phase gates , which are written in the standard basis as @xmath57 respectively .
adding the two - qubit controlled-@xmath1 ( or controlled - sign ) gate , @xmath58 completes the basis for clifford operations @xcite .
quantum - circuit notation @xcite is a pictorial method for representing the application of discrete operations to a quantum system . as with electrical circuits , repeated usage of a small number of simple , standard parts results in complex quantum circuits that are easier to understand and implement .
a typical component gate set contains a basis for clifford operations together with a single non - clifford operation . for brevity , we employ more clifford gates than are necessary to generate the clifford group , augmenting the gates of eqs .
( [ eq : hsdefinition ] ) and ( [ eq : czdefinition ] ) by the pauli matrices and @xmath59 .
any state can be expressed in terms of a quantum circuit that prepares it from some fiducial state , traditionally taken to be the one in which each qubit is initially in the state @xmath60 . to prepare any @xmath61-qubit stabilizer state from @xmath62
, it is sufficient to apply clifford gates , since , by definition of the clifford group , there exists a clifford operation that takes the stabilizer of the fiducial state to the stabilizer of the desired state under conjugation . applying this operation to the fiducial state yields a @xmath0 eigenstate of the stabilizers of the desired state , i.e. , the desired stabilizer state .
graph states can be written in a particularly simple form using quantum - circuit notation .
any graph state can be prepared using two layers of gates . in the first layer , @xmath63 is applied to each qubit . in the second layer , @xmath3 gates are applied between all pairs of qubits corresponding to connected nodes on the graph . to prepare an arbitrary stabilizer state
, it is sufficient to add a third layer that contains only @xmath63 and @xmath64 gates .
this is because any stabilizer state is equivalent to some graph state under local clifford operations @xcite .
any stabilizer state with a generator matrix of canonical form can be converted by local clifford operations to a state possessing a generator matrix of strict graph form . applying @xmath63 to the last @xmath65 qubits of the stabilizer state , where @xmath66 is the initial left rank of the canonical - form generator matrix , exchanges columns @xmath67 through @xmath4 in the generator matrix with columns @xmath68 through @xmath33 , so that a generator matrix as in eq .
( [ eq : canonicalstabilizer ] ) is transformed to @xmath69 the diagonal of @xmath49 in this graph - form generator matrix can then be stripped of @xmath70s without otherwise changing the generator matrix by applying @xmath64 to offending qubits .
the resulting generator matrix has the form of a graph state and corresponds to a stabilizer state that differs from that represented by eq .
( [ eq : canonicalstabilizer ] ) by at most a single @xmath63 or @xmath64 gate per qubit .
the close relationship between graph states and stabilizer states suggests the possibility of a graph - like representation of stabilizer states .
one approach to such a representation is simply to transform the generator matrix of a stabilizer state into strict graph form , draw the graph thereby obtained , and add decorations to each node indicating whether an @xmath63 or @xmath64 was applied to the corresponding qubit in the process of reaching strict graph form .
thus are stabilizer graphs constructed in this paper , where we choose to signal the application of hadamard gates by hollow ( unfilled ) nodes and the application of phase gates by loops .
the decision to represent @xmath64 gates by loops is motivated by the standard graph convention that a @xmath70 on the diagonal of an adjacency matrix denotes a loop on that node . and
@xmath71 on the left with the matching columns on the right , an operation that corresponds to applying a hadamard to qubits @xmath72 and @xmath71 . in graph form ,
the adjacency matrix is just the right half of the generator matrix .
loops arise from @xmath70s on the diagonal of the adjacency matrix , and hollow nodes are used to indicate which columns were exchanged between the right and left halves of the generator matrix to get the adjacency matrix .
notice that there are no edges between hollow nodes , nor are there any loops on hollow nodes.[fig : graph],width=321 ] the following steps provide a recipe for translating an arbitrary stabilizer generator matrix into a _ stabilizer graph _ : 1 . through row reduction and qubit swapping ,
transform the generator matrix into canonical form , as in eq .
( [ eq : canonicalstabilizer ] ) , keeping track during the process of how columns of the generator matrix map to qubits .
2 . draw the graph corresponding to the adjacency matrix @xmath73 including loops for @xmath70s on the diagonal .
3 . make solid the nodes corresponding to the rows and columns of the submatrix @xmath49 , and make hollow the nodes corresponding to the rows and columns of the submatrix @xmath42 .
notice that this procedure does not associate every combination of edges , loops , hollow nodes , and solid nodes with a stabilizer state . because the submatrix @xmath42 in eq .
( [ eq : baat0 ] ) contains only @xmath42s , hollow nodes never have loops , and there are no edges between hollow nodes .
we refer to stabilizer graphs having this property as _ reduced_. an example of a generator matrix and an associated reduced stabilizer graph is given in fig .
[ fig : graph ] .
it is important to note that the graphs in this paper are _ labeled graphs _ in that each node is associated with a particular qubit . swapping two qubits
is a physical operation that generally produces a different quantum state . in our graphs
gate can be described either by relabeling the corresponding nodes or by exchanging the nodes and all their decorations and connections while leaving the labeling the same . since the process of bringing the generator matrix into canonical form can involve swapping qubits
, we must keep track during this process of the correspondence between qubits and columns of the generator matrix and thus between these columns and the nodes of our graphs . just as a generator matrix contains no information about generator signs ,
so also are stabilizer graphs derived from generator matrices devoid of such information .
it is for this reason that we can be cavalier about whether @xmath64 or @xmath74 is used to convert a canonical generator matrix to strict graph form . in the absence of sign information ,
however , stabilizer graphs are best thought of as specifying an orthonormal basis rather than a single stabilizer state .
luckily , it is not hard to include sign information in the graph , as we show in the next subsection . having motivated a stabilizer - graph notation using generator matrices , we now turn to the quantum - circuit formalism to expand it .
the binary representation of stabilizers lacks a convenient way to keep track of generator signs , applied gates , and qubit swaps . since , for example , labeling is important , the column exchanges required to bring a generator matrix to canonical form must be tracked , perhaps by appending an extra row with qubit labels to the generator matrix .
details such as this are automatically dealt with when deriving stabilizer graphs from quantum circuits .
consider a quantum circuit consisting of three layers of gates applied to @xmath4 qubits , each initially in the state @xmath60 .
in the first layer , the hadamard gate , @xmath63 , is applied to each qubit . in the second
, controlled - sign gates , @xmath3 , are applied between various pairs of qubits .
finally , in the third layer , sign - flip gates , @xmath1 , phase gates , @xmath64 , and hadamard gates are applied ( in that order ) to various subsets of qubits .
we refer to such circuits as having _ graph form_.
the descriptor arises because a quantum circuit in this form can be depicted as a graph possessing three kinds of decoration .
in such a graph , two nodes are linked if a @xmath3 gate is applied between the qubits of the circuit corresponding to those nodes ; the various kinds of node decoration indicate the presence or absence of terminal @xmath1 , @xmath64 , and @xmath63 gates on the corresponding qubits .
it is the restrictions of such a representation that impel us to specify an order for the terminal gates , since @xmath75 .
the decoration corresponding to each gate is as follows : @xmath1 gates are denoted by a minus sign in the node , @xmath64 gates by a loop , and @xmath63 gates by a hollow ( as opposed to a solid ) node .
we refer to an arbitrary arrangement of solid and hollow nodes with loops , edges , and signs as a _ stabilizer - state graph _ or , more simply , as a _ stabilizer graph_. an example graph - form circuit and the corresponding stabilizer graph are given in fig .
[ fig : stabgraph ] . gates between qubits are transformed into links between nodes , terminating @xmath1 gates become negative signs , terminating @xmath64 gates result in loops , and terminating @xmath63 gates are denoted by hollow nodes .
nodes in ( b ) are labeled sequentially , beginning with the top node and moving clockwise .
[ fig : stabgraph],width=321 ] as might be guessed from our choice of decorations , the @xmath1 gates in a graph - form circuit specify the signs of the stabilizer generators .
the gates applied in the third layer of a graph - form circuit can be written as @xmath76 where @xmath77 , @xmath78 , and @xmath79 are binary variables taking on the values @xmath42 or @xmath70 . from eq .
( [ eq : thirdlayer ] ) it follows that the third layer of gates transforms a set of graph - state generators as in eq .
( [ eq : graphgenerators ] ) to the following stabilizer - state generators : @xmath80 equation ( [ eq : stabilizerstategenerators ] ) shows that the exclusive effect of each terminal @xmath1 operator is to flip the sign of a single stabilizer generator .
this can also be seen through circuit identities , since pushing a @xmath1 gate from the third layer of a graph - form quantum circuit to the beginning of the circuit merely transforms it to an @xmath81 gate ; flipping an input bit is equivalent to flipping the sign of the associated stabilizer since the stabilizer of @xmath60 is @xmath1 , which acquires a negative sign under conjugation by @xmath81 .
using either of these methods , it is clear that terminal @xmath1 gates , and hence the signs in stabilizer graphs , only impact the signs of the stabilizer generators , and can thus be omitted when these signs are thought to be unimportant .
modulo generator signs , the definition of stabilizer graphs given in the previous subsection and the definition given in this one are compatible .
the graph - form circuit and generator matrix associated with a stabilizer graph each specify the same stabilizer up to possible signs .
the method of deriving stabilizer graphs from generator matrices described in sec .
[ subsec : gengraphs ] , however , produces exclusively reduced stabilizer graphs , i.e. , stabilizer graphs satisfying the restriction that hollow nodes never have loops and never be connected to other hollow nodes . in terms of graph - form quantum circuits , this is the restriction that lines with terminating hadamard gates have no terminating @xmath64 gates and not be connected by @xmath3 gates .
the relative merits of stabilizer graphs and reduced stabilizer graphs are clarified in the following sections , particularly sec .
[ sec : equiv ] .
important roles are found for both .
it is frequently useful to consider the way in which stabilizer states transform under the application of clifford gates .
primarily , this is because clifford gates take stabilizer states to stabilizer states , a property that follows from their preservation of the pauli group .
this same property implies that the action of a clifford gate can be thought of as a transformation between the graphs representing the initial and final stabilizer states . in this section
we consider the transformations induced by local clifford gates .
the transformations induced by @xmath3 gates are discussed separately in the appendix .
we begin by introducing terminology for describing transformations on stabilizer graphs .
we use a number of terms , some adopted from graph theory and others invented for the task at hand . among those terms common to graph theory
are _ neighbors _ , _ complement _ , and _ local complement_. the neighbors of a node @xmath19 are those nodes connected to @xmath19 by edges ; we denote the set of neighbors of node @xmath19 by @xmath82 . in the transformation rules that follow , a loop does not count as an edge , so a node is never its own neighbor . complementing the edge between two nodes
removes the edge if one is present and adds one otherwise .
the local complement is performed by complementing a selection of edges , with the pattern of edges depending on whether local complementation is applied to a node or along an edge .
performing local complementation on a node complements the edges between all of the node s neighbors .
thus , local complementation at node @xmath19 transforms the adjacency matrix @xmath20 of a graph according to @xmath83 i.e. , it complements an edge if both nodes of the edge are neighbors of @xmath19 .
local complementation along an edge is equivalent to a sequence of local complementations on the _ decision nodes _ ,
i.e. , the nodes defining the edge .
the sequence is as follows : first perform local complementation on one of the decision nodes , then local complement on the other decision node , and finally local complement on the first decision node again .
this sequence transforms the adjacency matrix of a simple graph in the following way : @xmath84 where @xmath19 and @xmath14 are the decision nodes .
equation ( [ eq : threelc ] ) is symmetric in the two decision nodes , so it does not matter at which decision node local complementation is first performed .
additionally , since we do not consider self - loops to be edges , eq .
( [ eq : threelc ] ) can be applied to adjacency matrices with nonzero diagonal entries simply by ignoring those entries .
notice that an edge one of whose nodes is a decision node transforms according to @xmath85 . to describe the net effect of local complementation along an edge , it is helpful to define the _ decision neighborhood _ of a node as the intersection of its neighborhood with the decision nodes
. then local complementation along an edge can be summarized by three steps : 1 .
the edge between the decision nodes is left unchanged .
an edge one of whose nodes is a decision node is transferred from this decision node to the other .
an edge neither of whose nodes is a decision node is complemented if its end nodes have decision neighborhoods that are not empty and not identical . to these terms
we add _ flip _ and _ advance_. flip is used to describe the simple reversal of some binary property , such as the sign or the fill state ( color ) of a node .
advance refers specifically to an action on loops ; advancing generates a loop on nodes where there was not previously one , and it removes the loop and flips the sign on nodes where there was a loop .
its action mirrors the application of the phase gate , since @xmath86 .
an arbitrary local clifford operation can be decomposed into a product of @xmath63 , @xmath64 , and @xmath1 gates ( the @xmath1 is unnecessary , but convenient ) .
the effect of a local clifford operation on a stabilizer graph can thus be obtained by repeated application of the following six transformation rules . 1 .
applying @xmath63 to a node flips its fill .
2 . applying @xmath64 to a solid node advances its loop .
3 . applying @xmath64 to a hollow node without a loop performs local complementation on the node and advances the loops of its neighbors .
+ if the node has a negative sign , flip the signs of its neighbors as well .
4 . applying @xmath64 to a hollow node with a loop flips its fill , removes its loop , performs local complementation on it , and advances the loops of its neighbors .
+ if the node does not have a negative sign , flip the signs of its neighbors as well . 5 . applying @xmath1 to
a solid node flips its sign .
applying @xmath1 to a hollow node flips the signs of all of its neighbors .
if the node has a loop , its own sign is flipped as well .
these transformation rules can be derived from the circuit identities in fig .
[ fig : identproof ] , which rely on the basic circuit identities given in fig .
[ fig : basics ] . given an understanding of the relationship between circuits and graphs , transformation rules t1 , t2 , and t5 are trivial .
transformation rule t6 follows simply from fig .
[ fig : identproof](a , b ) .
transformation rules t3 and t4 derive from figs .
[ fig : identproof](c ) and ( d ) , respectively .
the transformation rules t1t6 do not generally take reduced stabilizer graphs to reduced stabilizer graphs . from sec .
[ subsec : gengraphs ] , however , we know that there exists a reduced stabilizer graph corresponding to each stabilizer state , so it is always possible to represent the effect of a local clifford operation as a mapping between reduced stabilizer graphs .
the appropriate transformation rules for reduced stabilizer graphs are listed below , excepting those for @xmath1 operations , which are identical to t5 and t6 .
applying @xmath63 to a solid node without a loop , which is only connected to other solid nodes , flips the fill of that node .
applying @xmath63 to a solid node with a loop , which is only connected to other solid nodes , performs local complementation on the node and advances the loops of its neighbors .
flip the node s sign , and if it now has a negative sign , flip the signs of its neighbors as well .
applying @xmath63 to a solid node without a loop , which is connected to a hollow node , flips the fill of the hollow node and performs local complementation along the edge connecting the nodes .
flip the sign of nodes connected to both the solid and hollow nodes .
if either of these two nodes has a negative sign , flip it and the signs of its current neighbors . applying @xmath63 to a solid node with a loop , which is connected to a hollow node ,
performs local complementation on the solid node and then on the hollow node .
then it removes the loop from the solid node , advances the loops of the solid node s current neighbors , and flips the fill of the hollow node .
flip the signs of nodes that were originally connected to both the solid and hollow nodes . if the originally solid node initially had a negative sign ,
flip it and the signs of its current neighbors , and if the originally hollow node initially had a negative sign , flip the signs of its current neighbors . applying @xmath63 to
a hollow node flips its fill .
applying @xmath64 to a solid node advances its loop .
applying @xmath64 to a hollow node performs local complementation on that node and advances the loops of its neighbors .
if the node has a negative sign , flip the signs of its neighbors as well . of these transformation rules , t(i ) , t(v ) ,
and t(vi ) are trivial , and t(vii ) is a rewrite of t3 . to prove the others requires results from sec .
[ sec : equiv ] , in particular , the equivalence rules in sec . [ subsec : circuitequiv ] .
specifically , t(ii ) is obtained by applying equivalence rule e1 , which gives an equivalent , but unreduced graph and then applying the hadamard , via rule t1 , which leaves a reduced graph . for t(iii ) , one first applies the hadamard , via rule t1 , and then uses equivalence rule e2 to convert to a reduced graph . in the case of t(iv ) , one applies the hadamard , using rule t1 , and then applies equivalence rule e1 , first to the originally solid node and then to the hollow node .
a key part of these transformations is the conversion of stabilizer graphs to reduced form , a process discussed in more detail in sec .
[ subsubsec : convert ] .
it is not hard to check that , in using rules t(iii ) and t(iv ) , any other hollow nodes that are connected to the originally solid node do not become connected and do not acquire loops , in accordance with the need to end up with a reduced graph .
as we have defined it , the mapping between a graph - form circuit and its corresponding stabilizer graph is one - to - one .
this does not imply , however , that each stabilizer state corresponds to a unique graph . on the contrary , an example of different graph - form circuits corresponding to the same stabilizer state
can be found in fig .
[ fig : identproof ] . applying an additional @xmath64 gate to the top qubit in fig .
[ fig : identproof](d ) makes the initial circuit identical to that in ( b ) , but the final circuits are quite different .
the two circuit identities thus define distinct transformation rules for applying a @xmath1 gate to a hollow node with a loop , thereby demonstrating the existence of multiple , equivalent graph - form circuits . for every way of obtaining a particular stabilizer state from a circuit in graph form
, there is an associated stabilizer graph . in this section
we examine the resulting equivalence classes of stabilizer graphs , first by presenting equivalence rules for full and reduced stabilizer graphs and then by introducing simplified graph - form - circuit equalities which we use to show that the equivalence rules given here are complete . applying either of the following two rules to a stabilizer graph yields an equivalent graph , i.e. , one which represents the same stabilizer state . 1 .
flip the fill of a node with a loop .
perform local complementation on the node , and advance the loops of its neighbors .
+ flip the node s sign , and if the node now has a negative sign , flip the signs of its neighbors as well .
2 . flip the fills of two connected nodes without loops , and local complement along the edge between them .
+ flip the signs of nodes connected to both of the two original nodes .
if either of the two original nodes has a negative sign , flip it and the signs of its current neighbors .
the first of these equivalence rules can be obtained by applying an additional @xmath64 gate to the top qubit in the identity of fig .
[ fig : identproof](d ) and equating the final circuit to the final circuit in fig .
[ fig : identproof](b ) . for the second rule ,
we need yet another circuit identity .
figure [ fig : transident](a ) shows how hadamards can be removed from a pair of connected qubits without @xmath64 gates .
figure [ fig : transident](b ) extends this identity to a demonstration of rule e2 .
the equivalence rules of the previous section can be reworked to yield equivalence rules for reduced stabilizer graphs .
the resulting equivalence rules are for a hollow node connected to a solid node with a loop , local complement on the solid node and then on the hollow node . then remove the loop from the solid node , advance the loops of its current neighbors , and flip the fills of both nodes . as for signs ,
follow the sequence in transformation rule t(iv ) .
for a hollow node connected to a solid node without a loop , local complement along the edge between them .
then flip the fills of both nodes . as for signs , follow the sequence in equivalence rule e2 .
there is a simple relationship between the two sets of equivalence rules .
equivalence rule e(ii ) is identical to e2 for the case that the two connected nodes have opposite fill .
equivalence rule e(i ) is simply rule e1 applied twice : first to the solid node with the loop and then once to the hollow node that has acquired a loop from the first application of e1 .
the second application of e1 is needed because the resulting graph is not reduced after one employment of the rule .
both of these equivalence rules can also be derived by applying two hadamards to a solid node , e(i ) handling the case in which the solid node has a loop and e(ii ) the case in which it does not .
thus equivalence rule e(i ) is simply transformation rule t(iv ) followed by use of rule t(i ) to apply a second hadamard to the originally solid node . likewise , e(ii ) is rule t(iii ) followed by t(i ) to apply a second hadamard to the originally solid node .
notice that both rules preserve the number of hollow nodes .
having described a set of rules for converting between equivalent stabilizer graphs , we show in this section that , in each case , the aforementioned rules generate the entire equivalence class of stabilizer graphs . the proof is divided into three parts .
the first part shows how to use rules e1 and e2 to convert an arbitrary stabilizer graph to an equivalent graph in reduced form .
the second part explains how an equivalence test for a pair of reduced stabilizer graphs can be simplified , using rules e(i ) and e(ii ) , to a special form .
finally , the third part proves that the graphs on the two sides of such a simplified equivalence test are equivalent only if they are trivially identical . taken as a whole
, this proves that the set of equivalence rules given in secs .
[ subsec : circuitequiv ] and [ subsec : redequiv ] is sufficient to convert ( reversibly ) between any two equivalent stabilizer graphs and thus that they generate all stabilizer graphs that are equivalent to the same stabilizer state . similarly , considering only the final two parts of the proof shows that the rules given in sec .
[ subsec : redequiv ] are sufficient to generate all reduced stabilizer graphs .
two features identify a stabilizer graph as reduced . in a reduced graph ,
hollow nodes never have loops , and hollow nodes are never connected to each other .
equivalence rule e1 can be used to convert looped nodes from hollow to solid . applying
rule e1 in this sort of situation can cause hollow nodes to acquire loops , but each application fills one hollow node of the graph , so the procedure will terminate in at most a number of repetitions equal to the number of hollow nodes in the graph .
similarly , connected hollow nodes in the resulting graph can be converted to solid nodes using the appropriate case of rule e2 .
once again , the process is guaranteed to terminate because the number of hollow nodes to which the rule might be applied decreases by two with each application of the rule .
concomitantly , the conversion of a stabilizer graph to an equivalent reduced graph never increases the number of hollow nodes .
equivalence testing for pairs of reduced graphs is facilitated by simplifying the equivalence such that nodes that are hollow only in the first graph never connect to nodes that are hollow only in the second .
this simplification can be accomplished by iterating the following process .
choose a pair of connected ( in either graph ) nodes @xmath87 and @xmath88 such that @xmath87 is hollow in one graph and @xmath88 is hollow in the other , and to the graph in which they are connected , apply the relevant reduced equivalence rule to the selected nodes . among other things ,
the equivalence operation reverses the fill of the two nodes it is applied to . since one node is hollow and the other solid , this preserves the total number of hollow nodes while yielding a node that is hollow in both graphs . because it is applied only to unpaired hollow nodes , subsequent uses of this equivalence rule do not disturb the newly paired hollow node .
consequently , this process also terminates in at most a number of repetitions equal to the number of hollow nodes in each of the graphs .
the two reduced graphs composing a simplified reduced - graph equivalence test are equivalent , i.e. , correspond to the same state , if and only if the graphs are identical . to see why this is so , we return to the graph - form quantum circuits discussed earlier .
in addition to the standard restrictions for reduced graphs , the circuits corresponding to the graphs in a simplified equivalence test have the following property : if in one of the circuits , a qubit with a terminal @xmath63 participates in a @xmath3 gate with a second qubit ( which can not have a terminal @xmath63 ) , then in the other circuit , it can not be true that the second qubit has a terminal @xmath63 and the first does not .
we prove the triviality of simplified reduced - graph equivalence testing by considering an arbitrary simplified reduced - graph equivalence test and showing that the two graphs must be identical if they are to correspond to the same state . in terms of unitaries ,
an arbitrary graph - form circuit equality can be written as @xmath89 where @xmath90 lists the pairs of qubits participating in @xmath3 gates and @xmath91 , @xmath92 , and @xmath93 are sets enumerating the qubits to which @xmath63 , @xmath1 , and @xmath64 gates are applied respectively .
the total number of qubits is denoted by @xmath61 and the subscripts @xmath94 and @xmath66 discriminate between the circuits on the left- and right - hand sides of the equation . in terms of these sets , a reduced - graph - form circuit satisfies the constraints @xmath95 and @xmath96 for all @xmath97 .
the circuits in simplified tests also satisfy @xmath98 for all @xmath99 and @xmath100 , where @xmath101 denotes the complement of @xmath91 , i.e. , the set of qubits to which @xmath63 is not applied .
suppose now that the two graphs have hollow nodes at different locations ; i.e. , at least one of the sets , @xmath102 and @xmath103 , is not empty . for specificity ,
let s say that @xmath102 is not empty .
in the language of circuits , this means that there exists a qubit @xmath87 that has a terminal @xmath63 on the left side of eq .
( [ eq : arbsimpgraphformcircequal ] ) , but not on the right side .
since qubit @xmath87 is part of a circuit for a reduced graph , it does not participate in @xmath3 gates with other qubits that possess terminal @xmath63 gates .
consequently , on the left side of eq .
( [ eq : arbsimpgraphformcircequal ] ) , the @xmath3 gates involving qubit @xmath87 can all be moved to the end of the circuit where they become @xmath104 gates with @xmath87 as the target . doing this and transferring the @xmath104 gates to the other side yields , @xmath105 where @xmath106 denotes the set of qubits that participate in @xmath3 gates with qubit @xmath87 on the left - hand side of eq .
( [ eq : arbsimpgraphformcircequal ] ) . because the original graph equality ( [ eq : arbsimpgraphformcircequal ] ) was simplified , @xmath107 and , by assumption , @xmath108 , so the @xmath104 gates and the terminal hadamards on the right side of eq .
( [ eq : modcircequal ] ) do not act on the same qubits .
thus we can commute the @xmath104 gates past the terminal hadamards .
moreover , we can then move the @xmath104 gates to the beginning of the circuit where they have no effect and can therefore be dropped .
during this migration , however , they generate a complicated menagerie of phases . the resulting expression for the right side of eq .
( [ eq : modcircequal ] ) is @xmath109 where @xmath110 is defined similarly to @xmath106 , @xmath111 and @xmath112 represents an indicator function , e.g. , @xmath113 equals @xmath70 if @xmath114 and @xmath42 otherwise .
it might appear that we have made things substantially worse by this rearrangement , but in an important sense , eq . (
[ eq : modcircequalright ] ) is now very simple with regard to qubit @xmath87 : @xmath63 is applied to qubit @xmath87 followed by a sequence of unitary gates all of which are diagonal in the standard basis .
this implies that there are an equal number of terms in the resultant state where qubit @xmath87 is in the state @xmath60 and the state @xmath115 . on the left side of eq .
( [ eq : modcircequal ] ) , however , the only gate remaining on qubit @xmath87 is the identity or an @xmath81 , depending on whether @xmath116 ; thus qubit @xmath87 is separable and is either in state @xmath60 or @xmath115 depending on whether @xmath116 .
consequently , our initial assumption that @xmath117 is incompatible with satisfying the equality .
the preceding discussion shows that two graphs composing a simplified equivalence test are equivalent only if they have hollow nodes in exactly the same locations .
given this constraint , the terminal hadamards can be canceled from both sides of eq .
( [ eq : arbsimpgraphformcircequal ] ) , giving @xmath118 the state after the initial hadamards is an equally weighted superposition of all the states in the standard basis .
the subsequent unitaries are diagonal in the standard basis , so they put various phases in front of the terms in the equal superposition . since a unitary is fully described by its action on a complete set of basis states , demanding equality term - by - term in eq .
( [ eq : paredsimpgraphformcircequal ] ) amounts to requiring that @xmath119 which is only satisfied when @xmath120 , @xmath121 , and @xmath122 .
thus , after simplification , the equivalence of pairs of reduced graphs is trivial to evaluate , since equivalence requires that the two graphs be identical .
an example of the entire process of testing graph equivalence is given in fig .
[ fig : equivalenceexample ] . an example which illustrates the circuit manipulations described algebraically in the text of this section
is given in fig .
[ fig : partthreeexample ] . as mentioned above
, our proof provides a constructive procedure for testing the equivalence of stabilizer graphs .
moreover , it shows that the set of equivalence rules given in sec .
[ subsec : circuitequiv ] is sufficient to convert between any equivalent stabilizer graphs and that the rules given in sec .
[ subsec : redequiv ] are sufficient to convert between any equivalent reduced stabilizer graphs . since the conversion of an arbitrary stabilizer graph to reduced form never increases the number of hollow nodes and the rules e(i ) and e(ii ) that convert among reduced graphs preserve the number of hollow nodes , we conclude that the reduced stabilizer graphs for a stabilizer state are those that have the least number of hollow nodes .
motivated by the relation between the graphs and the generator matrices associated with graph states , we extend the graph representation of states to encompass all stabilizer states .
these _ stabilizer graphs _ differ from the graphs employed by @xcite in that nodes can be either hollow or solid and can possess both loops and signs .
the additional decorations identify the local clifford operations that relate the desired stabilizer state to the graph state corresponding to the unadorned graph . imposing the restriction that the number of hollow nodes be minimal yields a subset of the stabilizer graphs which we term reduced .
reduced graphs follow naturally from the binary representation of the stabilizer formalism , while generic stabilizer graphs are more closely related to the quantum - circuit formalism . for graph states ,
reduced stabilizer graphs are identical to the standard representation of these states in terms of graphs . using circuit identities
, we derive a set of rules for transforming stabilizer graphs under the application of various local clifford gates . from this list , we abstract a similar set for transforming reduced stabilizer graphs . considering these transformation rules , particularly those for reduced stabilizer graphs , it becomes clear that the mapping between stabilizer states and stabilizer graphs is not one - to - one .
rules for converting between equivalent graphs are found , and we prove that the equivalence rules given are universal by developing a constructive procedure for testing the equivalence of any two stabilizer graphs .
this research was partly supported by army research office contract no .
w911nf-04 - 1 - 0242 and national science foundation grant no .
the transformation rules given in sec .
[ sec : transformations ] suffice to describe the effect of any local clifford operation on a stabilizer state . in order to complete the set of transformation rules for the clifford group , we include transformation rules for the @xmath3 gate here . in the interest of brevity
, we consider only reduced - stabilizer - graph transformations . transformation rules for general stabilizer graphs can be derived by first using the equivalence rules in sec .
[ subsec : circuitequiv ] to convert to an equivalent reduced graph and then applying the transformation rules below .
transformation rule t(viii ) is trivial since the @xmath3 gate simply commutes into layer two of the reduced - graph - form circuit .
the circuit identities needed to prove rules t(ix ) and t(x ) are given in fig .
[ fig : czident ] .
m. hein , w. dr , j. eisert , r. raussendorf , m. van den nest , and h .- j .
briegel , in _ entanglement in graph staes and its applications _ , proceedings of the international school of physics `` enrico fermi , '' course 162 , varenna , july 515 , 2005 , edited by g. casati , d. l. shepelyansky , p. zoller , and g. benenti ( ios press , amsterdam , 2006 ) , e - print arxiv : quant - ph/0602096 . | we introduce a graphical representation of stabilizer states and translate the action of clifford operators on stabilizer states into graph operations on the corresponding stabilizer - state graphs .
our stabilizer graphs are constructed of solid and hollow nodes , with ( undirected ) edges between nodes and with loops and signs attached to individual nodes .
we find that local clifford transformations are completely described in terms of local complementation on nodes and along edges , loop complementation , and change of node type or sign .
additionally , we show that a small set of equivalence rules generates all graphs corresponding to a given stabilizer state ; we do this by constructing an efficient procedure for testing the equality of any two stabilizer graphs . |
the study of the weak lensing distortion of background galaxies is a powerful tool for measuring the mass distribution of galaxy clusters .
it has long been recognized ( tyson et al .
1984 ) that the tidal field of a cluster acts by slightly modifying the images of distant , background galaxies . by measuring these small distortions the projected mass distribution of the lensing cluster can be derived in a model independent fashion ( kaiser & squires 1993 ) , in contrast with other mass estimators ( x - ray measurements of intra - cluster gas density and temperature , galaxy dynamics ) . to date
, there are approximately 20 clusters which have been well studied with lensing techniques ( see , e.g. , mellier 1999 for a review ) . in several cases the resulting masses
are claimed to be larger ( by a factor of two ) than those obtained using different methods . at present , it is not known if this discrepancy is due to biases and inaccuracies in the x - ray and dynamical estimates , or to the lensing data reduction method . as a matter of fact , it is not even clear whether there is a real discrepancy ( see wu et al .
1998 ; see also allen 1998 ) . for example , boehringer et al .
( 1998 ) have shown that different mass estimators are in excellent agreement when very good quality x - ray and optical data are used . in this paper , we study the mass distribution of the cluster ms1008@xmath01@xmath11224 ( in the following , for simplicity , ms1008 ) using high quality imaging data obtained during the science verification of fors1 at the vlt . the depth and seeing quality of these observations in four different bands ( bvri ) , make this dataset ideal for weak lensing mass reconstruction .
the large surface density of background galaxies in these images , as well as the sharpness of the psf and its small variations across the field , significantly improve the accuracy in measuring the distortion field from object ellipticities . as a result
, high signal - to - noise shear maps can be obtained .
this allows us to better investigate the effect of systematics involved in the mass inversion procedure , which in this case exceed statistical errors .
moreover , the availability of radial velocities for @xmath9 galaxies in ms1008 from the cnoc survey ( carlberg et al . 1996 ) , as well as x - ray spectroscopic observations , provide independent estimates of the total mass .
the paper is organized as follows . in sect .
2 we introduce the basic lensing relations and the notation used throughout the paper .
the observations and the imaging dataset are briefly described in sect .
section 4 is devoted to the weak lensing analysis of the cluster , where we describe in detail the mass reconstruction method and the results obtained . a virial analysis based on the cnoc redshift survey of ms1008 is provided in sect .
a comparison between the light and the mass distribution of ms1008 , and in particular its mass to light ratio , are discussed in sect .
finally , we draw the conclusions of this study in sect . 7 .
an independent lensing study using the same dataset has recently been presented by athreya et al .
( 2000 ) ; moreover , the depletion effect on this cluster has been recently investigated by mayen & soucail ( 2000 , preprint ) .
we briefly compare our results with these studies in the conclusions . throughout the paper we adopt @xmath10 , @xmath11 , @xmath12 .
in this section we briefly review the basic lensing relations .
we mostly follow the notation of seitz & schneider ( 1997 ) ( see also schneider et al .
1992 ) .
define @xmath13 to be the redshift of the lens ( in this case @xmath14 ) and @xmath15 the two - dimensional projected mass distribution of the lens , where @xmath16 is a two - dimensional vector representing a direction on the sky . for a source at redshift @xmath17
, we define the critical density , @xmath18 , which roughly describes the minimum density that a lens must have in order to produce multiple images from the source at redshift @xmath17 .
this quantity is given by @xmath19 here @xmath20 and @xmath21 are the angular diameter - distance from the observer to an object at redshift @xmath17 , and from the lens to the same object respectively . the ratio between the lens projected mass density , @xmath15 , and the critical density , @xmath22 , is defined as the dimensionless mass density @xmath23 .
this quantity characterizes the strength of the lens for sources at redshift @xmath17 ; in other words , all the lensing observables for galaxies at redshift @xmath17 depend only on @xmath24 . in general
, the source galaxies do not lie at a single redshift , @xmath17 , but are distributed with a some distribution @xmath25 ( here we define @xmath25 to be normalized so that @xmath26 ) .
this modifies the mean dimensionless mass density , @xmath27 , as @xmath28 note that the last equality holds because @xmath22 is defined to be infinity for foreground galaxies .
we define similarly a mean _ shear _
@xmath29 ( see , e.g. , seitz & schneider 1997 for a definition of the shear @xmath30 ) , which also depends on the last integral of eq .
( [ eq:<k > ] ) . in the weak lensing limit ,
i.e. in the limit @xmath31 for all @xmath16 and @xmath17 , all of the statistical lensing observables depend on @xmath32 and @xmath29 . in this analysis ,
the method employed to analyze galaxy images and infer the cluster is based on the weak lensing assumption . as shown by kaiser ( 1999 ) ,
such an assumption is justified for details on the mass distribution smaller than @xmath33 . for the cluster considered in this paper
this limit might not be valid in the central region , where the presence of a small arc suggests that the cluster is super - critical or at least nearly critical , and our results in this region should be interpreted with care . for simplicity
, we will refer in the following to @xmath34 and @xmath35 as averaged quantities , @xmath27 and @xmath36 .
ms1008 is a rich galaxy cluster drawn from the einstein medium sensitivity survey sample ( emss , gioia & luppino , 1994 ) at @xmath37 , with x - ray luminosity @xmath38 = 4.5 \times 10^{44}\ h_{50}^{-2 } \mbox { erg s}^{-1}$ ] .
ms1008 was also part of the cnoc survey ( carlberg et al .
1996 ) , whose data we will use extensively below .
this cluster was selected for the science verification ( sv ) of the first instruments on the vlt - antu ( ut1 ) to provide a deep multicolor imaging in b , v , r , i , j and k , _ the fors isaac cluster deep field _ ( see renzini , 1999 ) .
a detailed investigation of the mass distribution using gravitational lensing techniques was one of the main scientific goals of this sv programme .
observations were carried out in january 1999 .
fors ( focal reducer / low dispersion spectrograph , appenzeller et al .
1998 ) was used in standard resolution ( sr ) imaging mode , which provides a @xmath39 field of view with @xmath40 pixels . in order to avoid excessive bleeding and scattered light from two 11 magnitude stars in the northern part of the field ,
the fors mos mechanism was used in occulting mode , with two occulting bars on each side of the field designed to mask out the two bright stars at each position of the dithering pattern ( @xmath41 step ) .
this reduced the field of view by about @xmath42 , which , as discussed below , had a moderate impact on the weak lensing analysis .
we summarize in table 1 the main parameters of the optical imaging data .
near - ir images obtained with isaac , which cover a much smaller field of view , were not used in the present analysis .
the b - band image was obtained combining all the available frames .
the vri - band images utilized in our analysis were obtained coadding only the best seeing images of the entire set released by eso , resulting in final images with measured seeing of @xmath43@xmath44 , and depth very similar to that obtained coadding the whole set .
the flat - fielded images were registered and combined using procedures described in nonino et al .
( 1999 ) . for the coaddition , a modified version of the _ drizzling _ software originally developed for hst imaging ( fruchter & hook 1999 ) was used .
the photometric zero points are accurate at @xmath45 level , and the uncertainty on the zero - points do not have a significant impact on the lensing analysis .
.properties of optical images .
`` sb limit '' is the 1-@xmath46 surface brightness limit ; `` seeing '' is the mean fwhm obtained by fitting a moffat profile to a set of bright stars . [ cols="<,^,^,^ " , ] an alternative method to evaluate the radial profile is to use integrals of the shear ( fahlmann et al .
1994 ; squires & kaiser 1996 ) .
in particular , for a point located at polar coordinates @xmath47 we define the _ tangential shear _ as @xmath48 .
an estimator of the mass profile is given by @xmath49 which measures the mean dimensionless mass density @xmath50 inside a disk of radius @xmath51 relative to @xmath52 , the mean in an annulus of radii @xmath51 and @xmath53 .
we show @xmath54 in fig .
[ fig25-26.ps ] for each passband ; we used @xmath55 which defines the edge of the images .
the associated integral mass estimator , i.e. the quantity @xmath56 is shown in fig .
[ fig25-26.ps ] . in fig .
[ fig25-26.ps ] we also show in gray the same estimate as obtained independently by one of us ( g. squires ) using a slightly different technique ( see appendix ) .
the analysis of the internal dynamics of ms1008 , as traced by its member galaxies , has already been discussed in the literature by the cnoc collaboration .
the internal ( line - of - sight ) velocity dispersion for this cluster was originally measured by carlberg et al .
( 1996 ) , who found @xmath57 , based on the explicit background subtraction of the interloper contamination .
more recently , borgani et al . ( 2000 ) consistently found @xmath58 , based on the completely different method for removing interlopers , as described by girardi et al .
( 1998 , g98 hereafter ) .
carlberg et al . ( 1996 ) also estimated @xmath59 for the mass within the radius @xmath60 , which is defined as the radius encompassing an average overdensity of @xmath61 ( all the scales from the original papers have been suitably converted to those appropriate for the reference cosmological model assumed in our analysis ) .
lewis et al . ( 1999 ) estimated the mass within a radius @xmath62 from ` rosat ` hri x - ray data and found @xmath63 .
they also found this value to be consistent with that obtained by extrapolating to this smaller radius the virial mass estimate at @xmath64 with the nfw - like ( navarro et al .
1996 ) profile , fitted by carlberg et al .
( 1997 ) for the average density profiles of member galaxies for cnoc clusters . here
, we use the galaxy data ( redshift and positions ) as provided by yee et al .
( 1998 ) to determine the mass - profile of ms1008 from the virial analysis .
the procedure of the selection of member galaxies is the same of g98 .
in addition , we also reject emission line galaxies , since they are believed not to be fair tracers of the cluster internal dynamics ( e.g. , biviano et al .
1997 ) . from g98
we also adopt the procedure for estimating the velocity dispersion @xmath65 and the virial mass @xmath66 , for which we provide here only a short description . as for the velocity dispersion ,
we find @xmath67 km s@xmath68 , thus very close to previous estimates . in the case of a spherical system and under the assumption that mass follows galaxy distribution , it is possible to show ( e.g. , limber & mathews 1960 ; the & white 1986 ) that @xmath69 here @xmath70 is a term of surface pressure correction , which takes into account the fact that the radial component of the velocity dispersion does not vanish within the observational aperture radius ( carlberg et al .
1997 ) . in the following ,
we assume a @xmath71 correction at the virialization scale , which is typical for nearby and cnoc clusters ( g98 ; carlberg et al . 1996 ) .
the value of the projected virial radius @xmath72 depends on the relative ( projected ) positions of member galaxies within the sampled cluster region .
equation ( [ eq : mvir ] ) provides a reliable estimate of the virial mass only if galaxy redshifts are available over the whole virialized region of the cluster .
unfortunately this is not the case for ms1008 which is sampled only within about half of the virialized region ( cf.carlberg et al .
while this is not a serious problem for the estimate of @xmath65 , whose profile flattens well within the aperture radius ( borgani et al .
2000 ) , it may represent a limitation for @xmath73 , whose value increases with increasing clustercentric distance .
however , at each radius , an estimate of @xmath72 can be obtained by assuming a shape for the density profile .
for instance , g98 estimated @xmath72 by assuming a king - like profile , @xmath74^\alpha } \ ; , \ ] ] with core radius @xmath75 and shape @xmath76 [ cf .
their eq .
( 13 ) ] , which is the best fit to the average galaxy density profile for an extended sample of local clusters .
they also verified this estimate of @xmath73 to be reliable for clusters sampled out to the virialization scale . in this way
, we find a total virial mass which fully agrees with that given by carlberg et al .
( 1996 ) . *
* diamonds in fig .
[ fig : mass_vir ] show the projected mass profile obtained from the king - like profile of eq .
( [ eq : king ] ) .
figure [ fig : mass_vir ] also shows with the * * triangles the projected mass profiles obtained by scaling the original cnoc estimate with the nfw - like profile fitted by carlberg et al .
since no definitive conclusion has been reached at present about the most appropriate profile describing the cluster galaxy distribution ( see , e.g. , merritt & tremblay 1994 ; adami et al .
1998 ) , we prefer here to show the results for both such profiles .
althouh the two virial mass determinations coincide on scales comparable to the cluster virialized regions , the choice for the density profile affects the mass reconstruction at small radii . as a remarkable result
, we find an almost perfect concordance between our weak lensing mass and the @xmath77-ray mass by lewis et al .
( 1999 ) . * * we stress that the weak lensing profiles shown in fig . [ fig : mass_vir ] have been calculated by assuming a singular isothermal sphere for the cluster mass distribution ( we recall that this assumption is important when removing the mass - sheet degeneracy ) . as for the comparison with the virial mass reconstruction , a fairly good agreement is found at least for @xmath78 mpc , the virial mass being only slightly larger ( at about 1@xmath46 confidence level ) than that provided by the weak lensing reconstruction .
the difference becomes significant only at smaller radii , where , however , both the virial and the lensing masses become more uncertain .
these results indicate that the weak lensing mass mass reconstruction do provide results which agrees with other methods , based on the assumption of dynamical equilibrium , at least in cases when high - quality imaging data are available for a fairly relaxed cluster , like ms1008 ( see also allen 1998 ) .
to evaluate the mass to light ratio , we constructed a smoothed map of the light distribution in the b band , using a gaussian kernel with the same scale as the one used for the shear map .
the light density distribution is shown in fig . [
fig : mass - to - light ] , with overlaid mass contours , and the cluster galaxies used in our analysis .
galaxy fluxes were measured with the ` sextractor ` package ( bertin & arnouts 1996 ) , which ensures better object deblending in the cluster core with respect to ` imcat ` .
luminosities were computed by assuming a k - correction of @xmath79 magnitude in v band ( fukugita et al .
1995 ) , appropriate for ellipticals which dominate the cluster galaxy population .
luminosity evolution of early type galaxies can be neglected at this low redshift . in our cosmological model
the distance modulus is @xmath80 = 41.77 - 5 \log h_{50}$ ] .
the resulting rest - frame v - band luminosity summed over all the cluster galaxies is @xmath81 .
this estimate includes galaxies as faint as @xmath82 , hence the correction for incompleteness should be negligible .
carlberg et al . found @xmath83 @xmath84 in the gunn r band , at @xmath85 . using @xmath86 and @xmath87
, our estimate translates to @xmath88 at @xmath89 for @xmath90 ; hence it is in very good agreement with @xmath91 estimated by carlberg et al .
figure [ fig : mass - to - light ] clearly shows that , as commonly observed in clusters ( e.g. clowe et al .
1998 ) , the light roughly traces the mass .
the ne elongated morphology , clearly visible in the smoothed light distribution , is also shared by the intra - cluster gas , as visible in fig .
[ fig : mass - to - light ] , which shows the rosat - hri x - ray contours overlaid on the mass map ( grey scale )
. interestingly , the x - ray centroid is also displaced @xmath92 north of the cd , i.e. nearly coincident with the peak of the weak lensing signal .
the x - ray hri data are not deep enough to probe the outer regions of the cluster .
we also find some evidence for a drop of @xmath93 in the center possibly due to the presence of the over - luminous cd galaxy . in order to properly take into account the sheet degeneracy in the evaluation of the mass - to - light ratio
, we used an estimator similar to @xmath94 for the light distribution : @xmath95 where @xmath96 is v band luminosity averaged over a disk of radius @xmath97 .
thus , @xmath98 yields an unbiased estimate of the mass - to - light ratio ( assuming that the light traces the mass ) . using this method we obtained @xmath99 within a radius of @xmath100 , which compares favorably with the value found by hoekstra et al .
( 1998 ) , @xmath101 , for the cl 1358@xmath10262 , with redshift ( @xmath103 ) and x - ray luminosity ( @xmath104 = 5.5 \times 10^{44 } h_{50}^{-2 } \mbox{erg s}^{-1}$ ] ) very similar to ms1008 .
we have studied the weak lensing mass distribution of the cluster ms1008 using multicolor imaging data obtained during the science verification of fors1 on the vlt .
the depth , angular resolution , and image quality across the entire field of view make these images a rare and ideal dataset for weak lensing analysis . in addition
, the information on radial velocities of a large set of cluster galaxies provided by the cnoc survey allows the mass of this cluster to be derived with dynamical methods and compared with that reconstructed from the weak lensing .
these combined datasets allowed us to conduct a detailed study of the systematic effects involved in cluster mass estimates .
we have found psf distortions of fors1 to be moderate and hence easily removed with a low order polynomial fit .
the b through i multicolor information , as well as the redshift information from the cnoc catalog , have allowed us to efficiently separate the background galaxies from the cluster and foreground populations . approximately 8000 objects have been detected , corresponding to @xmath105 .
further selection lead to approximatively @xmath106 background galaxies with high signal - to - noise ratio which have been used for the determination of lensing shear field .
we have detected a weak lensing signal out to @xmath107 .
the projected mass within @xmath108 is measured in the range @xmath109 in the four filters , with typical statistical errors ( deriving from uncertainties in galaxy ellipticities ) of 5% .
we have discussed the impact of systematics in the weak lensing reconstruction of the mass distribution in physical units , which can be summarized as follows .
the removal of the mass - sheet degeneracy is inevitably model - dependent , inducing mass variations of @xmath110 at @xmath111 .
[ note that the estimate of this error is difficult and somewhat arbitrary since the error clearly depend on the `` range '' of possible models allowed . as a result ,
the estimate of the total error , given below , is also bound to be inaccurate .
we stress , however , that a similar problem exists for x - ray or virial mass estimate ( cf . ,
e.g. , the different predictions using king - like or nfw profiles if fig .
[ fig : mass_vir ] ) . ] by polluting the background galaxy population with a significant fraction of cluster galaxies , the total mass is biased low by @xmath112 .
however , different selections of background galaxies , as well as edge effects due to masked portions of the images , do affect the morphology of the shear maps .
we have not found any evidence for a substructure of the cluster core ; this in part is due to lack of resolution in the mass reconstruction , set by the smoothing scale used .
however , we have found the central region of the cluster to be elongated ns , with some evidence of a displacement of the projected mass peak respect to the cd galaxy , similarly to the x - ray emission .
cosmic variance is expected to affect the assumed redshift distribution of the background field galaxies , and hence the derived cluster mass . from a comparison between the hubble deep field north and south , cosmic variance
would introduce a systematic error of approximatively @xmath45 in the mass .
* * this can also be taken as a rough estimate of the error due to cosmic variance .
note also the main contribution to this error is from large scale structures rather than from poisson noise in the hdf - n and in our field . in other words , a simple error estimate which assumes that galaxies are uncorrelated would lead to an underestimate of the error .
altogether , the cluster light distribution traces the mass distribution remarkably well .
we have measured a mass - to - light ratio of @xmath113 at radii @xmath114 mpc .
we have also compared the weak lensing mass profile with that derived from a virial analysis of the cnoc redshift data , as well as with the x - ray mass .
the mass derived from x - ray observations ( lewis et al .
1999 ) is found in excellent agreement with our weak lensing determination at @xmath115 .
different approaches to estimate the cluster mass at the virialization scale produce very similar results . at scales
@xmath116 mpc , lensing and virial mass determinations can be compared .
the virial mass reconstruction at these small radii critically depends on the assumed density profile . using a king - like profile describing the galaxy distribution of local clusters ( g98
also eq .
( [ eq : king ] ) ) , we have obtained a mass which is more than @xmath8 away from the x - ray and weak - lensing mass .
assuming instead a nfw profile , as fitted by carlberg et al .
( 1997 ) to cnoc clusters , a much better agreement is obtained over the entire overlapping region .
note that the discrepancy at small radii ( @xmath117 ) between lensing and virial estimate can be explained by recalling that the lensing determination is affected by a gaussian smoothing of @xmath118 , as well as a departure from the weak lensing approximation .
the former effect leads to underestimate the mass in the core , while the latter will generally bias the central mass high .
both effects clearly vanish at large radii .
in addition , we notice that the weak lensing data alone can not discriminate between different mass profiles ( e.g. nfw versus isothermal model ) .
this analysis shows that the combination of depth and good imaging quality provided by vlt / fors allows us to obtain high @xmath119 mass maps via optimized weak lensing reconstruction methods . on the other hand ,
these data have the virtue of revealing important systematic errors in weak lensing analyses .
we find that the mass - sheet degeneracy dominates the budget of systematic errors and ultimately makes mass measurements via weak lensing model dependent .
the results obtained here are , at a first look , in good agreement with the mass reconstruction performed by athreya et al .
however , a detailed analysis of their study reveals that a different recipe to remove the mass - sheet degeneracy was used .
specifically , they set to zero the mass at large radii ( @xmath120@xmath121 ) .
had they used our ( model - dependent ) method to break the mass - sheet degeneracy , their estimate would have been a factor of two higher .
therefore , there is a discrepancy of the same factor between the two analysis which , at present , we are unable to explain .
the authors would like to thank the ut1 science verification team at eso for the preparation of this unique data set and its release to the community .
we are particularly grateful to the vlt / ut1 commissioning team for carrying out the observations .
we also thank tom broadhurst , tomas erben , and peter schneider for useful discussions .
support for this work was provided by nasa through hubble fellowship grant no .
hf-01114.01 - 98a from the space telescope science institute , which is operated by the association of universities for research in astronomy , incorporated , under nasa contract nas5 - 26555 .
in the second analysis of the data , we followed the procedure used by hoekstra et al .
the anisotropic part of the psf is removed on an object - by - object basis by correcting each object s ellipticity via @xmath122 where starred quantities refer to parameters measured for stellar images . in this analysis , we calculated the stellar ellipticities and polarizability employing a matched radius for the gaussian weight function as for the object in question .
we fit a second order polynomial to account for spatial variations in the psf .
we next determine the shear from the anisotropy - corrected galaxy polarizations .
both the ( now effectively isotropic ) psf , and the circular weight function , tend to make objects rounder .
these effects may be corrected for using the ` pre - seeing shear polarizability ' @xmath123 introduced by luppino & kaiser ( 1997 ) , where the @xmath124 subscript again refers to quantities determined from the objects classified as stars .
operationally , to determine @xmath125 , we split the faint galaxy catalogs into six bins based on size @xmath126 , with spacing @xmath127 pixels .
we reject galaxies with very large , small , or negative values of @xmath128 , and @xmath129 ( the off - diagonal terms typically being very small ) .
the rationale for this cut is that the calculation of the shear and smear polarizabilities for these objects failed , typically either because the objects themselves were very faint and near the detection threshold , or due to the presence of very nearby object contaminating the aperture over which the calculations were performed . in each @xmath126 bin
, we calculate the _ median _ galaxy polarizabilities , @xmath130 and @xmath131 , and reanalyze the stellar catalog , resetting the value of the stellar @xmath126 to the bin size , and analyzing with a gaussian with scale @xmath126 and an aperture three times this radius . | we present an in - depth weak lensing analysis of the cluster ms1008@xmath01@xmath11224 based on deep multicolor imaging obtained during the science verification of fors1 at the vlt .
the image quality ( half arcsec seeing ) and depth of the vlt images allow the shear signal to be mapped with high signal - to - noise and to be traced out to @xmath2 , near the edge of the @xmath3 field of view . using bvri color information , as well as 81 redshifts in the field from the cnoc survey
, background galaxies can be effectively separated from cluster and foreground objects .
psf distorsions are found to be moderate across the fors images and thus easily removed . due to the small statistical errors in the mass reconstruction
, this dataset provides a testing ground where several systematic effects ( e.g. mass - sheet degeneracy , redshift distribution of the background sources , cluster galaxy contamination ) , which are involved in the weak lensing analysis , can be quantified .
several methods are used to remove the mass - sheet degeneracy which is found to dominate the systematic error budget .
we measure a lower limit to the mass of @xmath4 within @xmath5 and a `` total '' mass of @xmath6 by fitting a softened isothermal sphere .
we find the mass distribution fairly uniform , with no significant substructures , in agreement with the virial analysis .
the availability of the cnoc redshift data and x - ray observations on this cluster allow a comparison of different determinations of the mass radial profile .
we find the lensing and x - ray measurements in excellent agreement , while the mass derived from the virial analysis is marginally ( @xmath7@xmath8 ) in agreement at radii where both methods are reliable .
this analysis underscores the importance of systematics in the mass determination of clusters , particularly when such a high quality dataset is not available or in similar studies at higher redshifts . |
one of the most direct methods to establish the coupling of an electronic property with the dynamics of the underlying lattice is the isotope effect ( ie ) , @xmath5 , which probes the dependence of the electronic quantity @xmath6 on the ion mass @xmath7 .
for example , the observation of a large ie on the superconducting transition temperature @xmath8 has been an important evidence for a phonon - mediated pairing mechanism in conventional low - temperature superconductors . in the weak - coupling bcs model ,
the superconducting critical temperature is proportional to the debye phonon frequency which scales as @xmath9 .
the resulting ie on @xmath8 , @xmath10 , has been considered as one of the most important confirmation of the bcs theory .
the vanishingly small values of @xmath11 in optimally doped high-@xmath8 copper oxides superconductors has initially induced a large part of the scientific community to believe that the pairing mechanism in these compounds was mainly of electronic origin , and oriented the theoretical research towards purely electronic models , relegating the lattice to a secondary role .
later experimental investigations showed however that small values of @xmath11 are peculiar of the optimal doping , whereas sizable values of @xmath11 , even larger than the bcs prediction , were reported in the underdoped phase.@xcite furthermore , the improvement of the experimental accuracy has allowed recently to establish the presence of isotopic shifts also in several physical quantities different from the superconducting @xmath8 .
sizable isotope effects ( ies ) have been indeed found on the zero temperature penetration depth @xmath12,@xcite on the knight shift,@xcite on the nuclear quadrupole resonance,@xcite on the pseudogap temperature,@xcite and on the angular resolved photoemission spectra.@xcite these findings are particularly remarkable since the migdal - eliashberg theory of the electron - phonon interaction predicts vanishingly small ies other than @xmath11 .
for example , according to the bcs description of the superconducting state , the zero temperature penetration depth is given by @xmath13 , where @xmath14 is the density of the condensate and @xmath15 is the effective electron mass .
according to the migdal - eliashberg theory , @xmath16 , where @xmath17 is the bare band mass and @xmath18 is the electron - phonon coupling constant . since @xmath18 is independent of the ion mass , the isotope effect on the penetration depth
is expected to be zero , in contrast with refs . .
the completely anomalous set of isotope dependences signals a relevant electron - phonon coupling , but at the same time highlights that the interplay with strong correlations and possibly with other features of the cuprates needs to be taken into account .
different theoretical models have been proposed to account for these anomalous ies.@xcite the analysis has been mainly concentrated on purely electron - phonon narrow - band systems , as the holstein model which however neglects possible effect of the anisotropy of the electron - phonon interaction considered in ref . .
the ies on @xmath15 in interacting electron - phonon systems in the absence of electron - electron interactions have been investigated in details in refs . .
more recently , the ability of the ies in revealing different physical regimes of electron - phonon systems has been discussed in ref . where remarkable anomalous ies on @xmath15 and on the dressed phonon frequency @xmath19 were reported as the system enters in a polaronic regime . in order to focus on the metallic properties and to clarify the origin of these anomalous ies ,
however , a spinless holstein model in the dmft approximation was considered there , which enforces the metallic character in the whole space of parameters . in the opposite case of strong correlations
the electron - phonon interaction in the weak - intermediate limit has been treated for instance by means of an adiabatic theory on small clusters,@xcite or of coherent potential approximation.@xcite in this case the ie reflects in a smooth modification of high energy features such as the hubbard bands . in the strongly correlated antiferromagnetic phase ies on the single hole spectral properties have been also analyzed with diagrammatic quantum monte carlo techniques.@xcite the scope of this paper is to provide a nonperturbative analysis of the isotope effects on different observables in a regime in which strong correlation and intermediate / strong electron - phonon interaction coexist . to this aim
we employ the dynamical mean - field technique ( dmft ) , which is powerful tool to investigate the nonperturbative regimes of strongly correlated and electron - phonon systems . in particular ,
in this paper we consider the metallic regime of the hubbard - holstein model , deliberately excluding broken - symmetry phases .
we also limit ourselves to the half - filled systems , where in dmft the correlation - driven mott transition or the electron - phonon driven pairing transition take place as one or the other coupling becomes large , and we follow the behavior of the isotope effects when these phase boundaries are approached .
it is well known that dmft becomes exact only in the limit of infinite coordination number , and it has to be viewed as an approximate method in finite dimensions .
yet , dmft allows us to access the full local quantum dynamics , and to deal with many different energy scales .
moreover the assumption of a local self - energy , implicit in the dmft , makes this technique not suitable to investigate the momentum dependence of the physical quantities .
this is for instance the case of frlich - like electron - phonon hamiltonians and of the long - range coulomb interaction , which gives rise to important screening effects here neglected.@xcite this is also the case of the momentum dependence of the electron - phonon interaction induced by the electronic correlation itself@xcite , which can not be properly take into account within the context of the dmft .
a general belief is that the electronic correlation competes with the electron - phonon interaction leading to a reduction of the phonon effects on the electronic properties with a consequent decrease of the magnitude of the possible ies .
we show that in particular regimes , close to phase boundaries , the hubbard repulsion can actually _ enhance _ the dependence of many physical quantities on the phonon frequency , yielding an increase of the ies .
indeed the isotope effects turns out to be extremely sensitive to the closeness of metal - insulator transitions , and emerge as a tool to investigate the physics underlying the different regimes of the phase diagram .
this paper is organized as follows . in sec .
[ approach ] we introduce the hubbard - holstein model and the dynamical mean field method . in sec .
[ s - weak ] and sec .
[ s - strong ] we report our results for the weakly correlated and strongly correlated regimes , respectively , while a study of the ies as a function of the hubbard-@xmath2 repulsion is the subject of sec .
[ u - dependence ] .
the conclusions are summarized in sec .
the hubbard - holstein model is defined by the hamiltonian : @xmath20 where @xmath3 is the hopping amplitude between nearest neighboring sites , @xmath2 is the hubbard electron - electron repulsion , @xmath21 is the electron - phonon matrix element , and @xmath22 the einstein phonon frequency .
we can identify three independent dimensionless parameters , the electron - phonon coupling constant @xmath23 , the adiabatic ratio @xmath24 , which parametrizes the role of the lattice quantum fluctuations , and the dimensionless hubbard repulsion @xmath25 .
it is important to note that @xmath26 and @xmath27 , where @xmath7 is the ionic mass , so that the only physical quantity which depends on @xmath7 is the adiabatic ratio @xmath28 .
finite ies stem thus from an explicit dependence on @xmath28 and they are hence a direct signal of the role of the lattice quantum fluctuations . the locality of the interaction terms makes the hubbard - holstein model particularly suitable to be investigated by dynamical mean - field theory.@xcite in this approach the spatial fluctuations are frozen , i.e. , all the lattice sites are assumed to be equivalent , but the local electron and phonon dynamics are fully accounted for . in this sense
the method generalizes to a quantum framework the classical mean - field theories , and it becomes exact in the infinite coordination limit .
hence the lattice model is mapped onto a quantum impurity model in which an interacting site is embedded into a non - interacting bath , whose spectral function has to be self - consistently determined . in our case
, the impurity site displays both coulomb repulsion and a coupling to a local phonon .
the self - consistency equation contains all the information about the original lattice model through the bare density of states .
as often , here we consider an infinite coordination bethe lattice with half - bandwidth @xmath3,@xcite which reproduces the finite bandwidth of finite - dimensional systems .
for the normal paramagnetic phase , the self - consistency condition on the bethe lattice reads @xmath29 where @xmath30 is the green s function on the impurity site , which corresponds to the local component of the lattice green s function , and @xmath31 is the bath hybridization function . within dmft
the electron and phonon self - energies are local , i.e. , momentum independent . to solve the impurity model we use the exact diagonalization method,@xcite in which the dynamical dmft bath is described in terms of a finite set @xmath32 of anderson impurity levels , while a finite cut - off is imposed on the total number @xmath33 of the phononic hilbert space .
typical values we considered are @xmath34 , and @xmath35 .
both these quantities were varied to check the effective stability of the results .
the self - consistent dmft set of equations was solved in the matsubara space where a fictitious temperature @xmath36 plays the role of an energy cut - off .
particularly small value of @xmath36 up to @xmath37 were needed to ensure a high accuracy of our results . in this paper
we investigate unconventional ies on both single - particle and two - particle quantities .
in particular , concerning the one - particle properties , we focus on the quasi - particle spectral weight @xmath0 , which in dmft coincides with the inverse of the effective electronic mass @xmath38 due to the momentum independence of the self - energy @xmath39 @xmath40 and on the renormalized phonon frequency @xmath41 obtained as@xcite @xmath42 in the above expressions , @xmath43 is the electronic self - energy and @xmath44 is the phonon propagator @xmath45 , where @xmath46 is the dimensionless lattice coordinate operator .
@xmath47 and @xmath48 are respectively fermionic @xmath49 and bosonic @xmath50 matsubara frequencies . in practice @xmath0
is computed by linearly extrapolating from the first matsubara frequency .
the knowledge of the renormalized phonon frequency @xmath19 allows us to evaluate also the static component of the _ local _ charge susceptibility @xmath51 , where @xmath52 , @xmath53 being the density operator on the impurity .
it is indeed easy to prove that @xmath54 .
we evaluate also the local static spin susceptibility @xmath55 for the spin response function @xmath56 , where @xmath57 is the @xmath58-component of the local electron spin ( @xmath59 being the pauli matrix @xmath60 ) .
finally , we consider also the local lattice probability distribution function @xmath61 , where @xmath62 is the eigenstate of the lattice coordinate operator @xmath63 . in our exact diagonalization approach
we can evaluate @xmath64 , where @xmath65 is the ground state , @xmath66 are the eigenstates of the harmonic oscillator and @xmath67 the corresponding eigenfunctions . despite its simplicity
, the hubbard - holstein model has quite a rich phase diagram , in which different phases and regimes appear . in order to have a better comparison between the different regimes
, we organize our discussion according to degree of correlation . in the next section we focus on the weakly correlated regime of the holstein - hubbard model , where the presence of the on - site hubbard repulsion changes only quantitatively the results of the pure holstein spinful model , but it does not give rise to new features .
we then discuss in section [ s - strong ] the regime of strong electronic correlation where the interplay between electron - phonon and electron - electron interaction induces qualitative changes on the behavior of the ies .
already in the absence of correlation , the introduction of the spin degeneracy introduces additional physics with respect to the spinless system , whose isotope effects have been studied in ref . .
in the spinless case indeed the polaron crossover is not associated with a metal - insulator transition ( mit ) which is only asymptotically approached for @xmath68 .
the persistence of a metallic character in the polaronic phase is reflected in the finite value of both the quasi - particle weight and the renormalized phonon frequency.@xcite on the other hand in the spinful case the residual attractive interaction between the electronic or polaronic charges leads to the possibility to form a insulating phase of localized pairs which in the adiabatic regime ( @xmath69 ) are accompanied by strong lattice deformation ( bipolarons).@xcite this occurs within the dmft approximation for @xmath18 larger than a finite critical value where the quasi - particle weight vanishes at @xmath70 signaling a pair - mit.@xcite in the adiabatic limit ( @xmath71 ) the pair - mit has a precursor in the phonon softening which occurs at @xmath72.@xcite as the bare phonon frequency is increased , the phonon softening tends to occur more closely to the pair - mit.@xcite in fig .
[ f - uweak ] we show the quasi - particle spectral weight @xmath0 , the renormalized phonon frequency @xmath19 , the local static spin susceptibility @xmath73 and the local static charge susceptibility @xmath74 as functions of the electron - phonon coupling @xmath18 for different weak - intermediate values of the hubbard repulsion @xmath2 and for adiabatic ratios @xmath75 .
@xmath76 and @xmath77 represent respectively the local spin and charge susceptibility in the absence of electron - phonon and hubbard interaction . as mentioned above
, the main difference here with respect to the spinless holstein model is the possibility to form local bipolaronic pairs of electrons with opposite spin .
this leads to a metal - insulator transition at a finite electron - phonon coupling @xmath70 where @xmath78 .
the metal - insulator transition is accompanied by a softening of the phonon frequency @xmath79 for a value @xmath72 of the electron - phonon very close to @xmath70 , followed by a sudden increase for @xmath80.@xcite the approach to the bipolaronic metal - insulator transition is reflected also in the two - particle quantities @xmath73 and @xmath74 .
in particular , the increase in the number of singlet local electron pairs close to the transition leads to a suppression of the local spin susceptibility and to an enhancement of the local charge susceptibility , so that @xmath73 has a sharp drop whereas @xmath74 nearly diverges at @xmath72 . in this context the hubbard repulsion gives rise to a further reduction of metallic properties at small @xmath18 ( decrease of @xmath0 ) while , on the other hand , it disfavors double occupancy leading to a shift to larger @xmath18 of the critical values of the el - ph coupling and to an increase ( decrease ) of the spin ( charge ) susceptibility .
finally , increasing the quantum lattice fluctuations ( @xmath81 , right panels ) leads to a further increase of the critical electron - phonon couplings . in fig .
[ f - isouweak ] we show the corresponding isotope coefficients @xmath82 where @xmath83 is @xmath0 , @xmath19 , @xmath73 and @xmath74 .
we have chosen to compute the ies using a finite value @xmath84 .
the behavior of ies approaching @xmath70 and @xmath72 is depicted in fig .
[ f - isouweak ] .
the most striking effect is the near _ divergence _ of the isotope coefficients as the system approaches the pairing metal - insulator transition . as a matter of fact , while @xmath85 has a true divergence as @xmath86 , for the isotope effects on the other quantities the divergence is cut - off by the not complete suppression of charge fluctuations . from the practical point of view
however both the mit and bipolaronic instabilities are accompanied by a huge increase of the magnitude of the anomalous isotope effects .
the key point to understand this issue is that an isotopic shift of @xmath22 is reflected in a finite shift to higher @xmath18 of both the bipolaronic mit ( @xmath70 ) _ and _ the phonon softening ( @xmath72 ) .
it obviously descends from eq .
( [ def : alpha ] ) that the divergence of a quantity is accompanied by a divergence to @xmath87 of the corresponding isotope coefficient @xmath88 , while a positively divergent isotope coefficient is associated to a vanishing observable .
according this view we can understand the behaviors of fig .
[ f - isouweak ] .
namely we find that @xmath89 as @xmath78 at @xmath70 and @xmath90 nearly diverges as @xmath73 rapidly drops at @xmath70 . in similar way the near vanishing of @xmath19 and the near divergence of @xmath74 , at @xmath72 are reflected in @xmath91 and @xmath92 .
the following drop of @xmath93 and the jump of @xmath94 for @xmath95 reflect the sharp behavior shown in fig.[f - uweak ] of the corresponding quantities .
increasing @xmath2 leads to an increase of @xmath70 , while increasing @xmath22 determines an overall smoothing of these features .
an interesting point to be noted here is the non monotonic behavior of the ie on the spin susceptibility @xmath73 , which shows a negative sign at small @xmath18 before approaching a near divergence towards @xmath96 close to the metal - insulator transition .
on the contrary @xmath94 remains negative up to @xmath72 .
the negative sign of @xmath97 and @xmath94 for small @xmath18 can be understood in a perturbative framework .
the electron - phonon interaction introduces a positive stoner - like contribution to the effective spin susceptibility : @xmath98 where @xmath99 arises from the electron - electron repulsion and the electron - phonon term @xmath100 takes into account the explicitly isotope dependence due to the finite bandwidth and vertex corrections arising from the electron - phonon interaction.@xcite in the same rpa fashion the charge susceptibility reads @xmath101 where @xmath102 .
thus at a perturbative level the only source of ie in @xmath74 and @xmath73 comes from @xmath103 , which can be shown to be the same as in both quantities leading to the _ same _ negative ie .
the sign of the isotope coefficient on the spin susceptibility reflects thus two different physical regimes where lattice fluctuations triggered by @xmath22 play a different role .
namely , in the perturbative small @xmath18 regime lattice fluctuations disfavor single spin occupancy ( fig .
[ f - p ] ) , leading to a decrease of @xmath73 and to a negative isotope coefficient .
on the other hand the physics in the @xmath104 region is dominated by the proximity to the bipolaronic mit . in this context lattice fluctuations
relax the phonon - induced electron trapping favoring the unpaired spin singlet states . along this line
the sign of the isotope coefficient on @xmath73 can be thus used as a useful trademark to characterized the different physical regimes .
the ability of the ie to characterize the physical properties of different regimes will be even more remarkable in the strongly correlated regime .
in the previous section we have seen that for small @xmath2 the physical properties of the system are qualitatively similar to those of the pure holstein model , where turning on a small coulomb repulsion leads essentially only to an increase of the critical value of @xmath18 for bipolaron pairing .
the situation is significantly changed in the strongly correlated regime in which the hubbard repulsion exceeds the electron - phonon coupling . in this case a more reasonable starting point is the physics of the pure hubbard model whose metallic region , close to the mott transition , can be described in terms of a narrow resonance at fermi level accompanied by the two high - energy hubbard subbands . in this framework
the electron - phonon interaction has two opposite effects : on one hand it provides a further source of scattering which would decrease the quasi - particle scattering weight @xmath0 ; on the other hand it acts as a effective attraction between the electrons , competing with the hubbard repulsion.@xcite in the strongly correlated regime , when the reduction of the effective repulsion prevails over the electron - phonon renormalization of the quasi - particle properties , increasing @xmath18 leads to an enhancement of @xmath0.@xcite another interesting feature of the hubbard - holstein model pointed out in ref . in the strongly correlated regime is the change of the order of the bipolaronic mit at @xmath105 .
the bipolaronic transition , which is continuous ( second order ) for @xmath106 , becomes indeed discontinuous ( first order ) for @xmath107 . in this latter regime
all the observable have thus a jump at @xmath70 .
the evolution from a second order to a first order transition has been discussed in terms of a landau - ginzburg theory where the first / second order transition is driven by the local lattice configuration.@xcite from a numerical point of view , however , it is hard to distinguish , on the basis of observables like @xmath0 or @xmath19 a first order transition from a sharp second order one . we show that the analysis of the phonon ie is a powerful tool to distinguish in a very clear way the first order or the second order character of the bipolaronic mit .
this is even more interesting since , as we are going to show , the first order transition is not related to a phonon mechanism while it is probably of a electronic origin . in fig .
[ f - ustrong ] we show @xmath0 , @xmath19 , @xmath73 and @xmath74 as function of @xmath18 in the regime of strong correlation for @xmath105 and for @xmath108 . as mentioned above , in this strongly correlated regime ,
increasing @xmath18 , as long as the system does not reach a bipolaronic regime , results in a reduction of the effective hubbard repulsion and in an increase of @xmath0 .
it is however interesting to note that the behavior of the other quantities , namely @xmath19 , @xmath73 and @xmath74 , as function of @xmath18 and @xmath2 does not change qualitatively with respect to the weakly correlated case .
concerning the character of the bipolaronic mit , we find that the transition is always of the second order for any @xmath2 at @xmath108 . on the other hand , due to the lack of precursor effects in the metallic phase , it is hard to predict the character of the transition at @xmath105 .
the corresponding ies on the quantities @xmath0 , @xmath19 , @xmath73 and @xmath74 is reported in fig .
[ f - isoustrong ] .
we note a qualitatively different behavior of the isotope coefficients in the strongly correlated regime at @xmath105 as compared with the weak @xmath2 regime and ( partially ) with the case @xmath81 .
let us discuss for the moment the highly correlated cases @xmath109 .
most striking is the behavior of the phonon and of the charge susceptibility isotope coefficient which shows a trend diametrically opposite to the weak-@xmath2 case , namely a marked _ reduction _ ( increase ) of @xmath93 ( @xmath94 ) to be compared with the divergences @xmath110 , @xmath111 for small @xmath2 . among the other anomalous features we note a typical downturn of @xmath85 as function of @xmath18 , which is less marked as @xmath2 increases . on the contrary , the upturn of the isotope coefficient on the spin susceptibility which is evident at small @xmath2 is here absent . quite interesting
is also the @xmath112 case which is very close to the value where the bipolaron mit changes from the second order to the first order . in this case
the ies show similar trends ( @xmath113 , @xmath114 , downturn of @xmath85 ) as for the first order cases in a large range of @xmath18 , with a final upturn ( downturn for @xmath94 ) towards the characteristic divergences of the second order .
this trend is clearly visible also at @xmath81 ( although the bipolaronic transition is always of the second order ) pointing out an partial underlying tendency towards a first order instability .
the larger amount of the lattice fluctuations in this case as compared with @xmath105 prevents however the occurrence of the first order instability and enforces the second order character .
as we are going to show , the physical origin of these anomalous behaviors can be related to different phonon regimes , although they do not rule directly the change between the first and second order character of the transition . in order to clarify in more details this issue
we evaluate the local lattice probability distribution function ( pdf ) @xmath115 which provides information about the lattice wavefunction of the ground state . in fig .
[ f - pxvx ] ( upper panel ) we show the pdf for two representative cases characterized by significant lattice distortions induced by the electron - phonon interaction , resulting in a non - gaussian @xmath115 . in the strictly adiabatic case , polaronic effects for instance
are reflected by a bimodal shape of @xmath115 , where the two maxima are associated to different electron occupancies ( e.g. , in spinless systems @xmath116 and @xmath117 , in spinful systems @xmath116 and @xmath118 ) .
the broadness of these structures induced by quantum lattice fluctuations however can make hard to resolve the effective extent of the polaronic lattice distortions . to overcome this problem we extract an `` effective '' lattice potential @xmath119 by solving the inverse schrdinger problem @xmath120\bar{\psi}(x)=\omega_0\bar{\psi}(x),\ ] ] where @xmath121 and @xmath7 is the atomic mass .
[ f - pxvx ] shows the lattice distribution function @xmath115 and the corresponding lattice potential @xmath119 for two representative cases where the multi - valley structure of @xmath119 is not reflected in the shape of @xmath115 .
we stress that the above defined @xmath119 reproduces the physical lattice potential only in the strict adiabatic limit @xmath122 .
it can be considered however to be still a valid approximation at finite and small @xmath28 providing thus useful insights for the true ( unknown ) lattice potential . in fig .
[ f - psvx ] we report the evolution of the lattice potential @xmath119 as function of @xmath18 for three representative cases of the hubbard repulsion , @xmath123 , and for @xmath105 .
all the shown @xmath18 s belong to the metallic phase . the largest value of @xmath18 , in each panel of the left side of fig .
[ f - psvx ] , corresponds to the mit .
the first case ( panel a ) shows , in agreement with a second - order transition scenario , a continuous flattening of the lattice potential as @xmath18 increases .
this picture shows in the clearest way the lattice instability at @xmath124 , where the quadratic term of @xmath125 vanishes and the renormalized phonon frequency @xmath79 . for @xmath126 the systems
is still metallic and a double - well structure of the lattice potential is developing .
this corresponds to the phonon hardening back shown in fig .
[ f - uweak ] .
the electronic metal - insulator transition occurs for @xmath127 when the double - well potential is already established and the mit is caused by the vanishing of coherent tunneling between the two minima .
a crucial role is played by the zero point phonon energy @xmath128 ( horizontal lines in panel b ) which favors tunneling processes . in particular , in the latter situation where the double - well is already formed , an increase of @xmath129 , due to an isotope shift , further enhances the metallic character , enhancing @xmath0 and resulting in a positive isotope coefficient on @xmath0 . at the same time
the system , increasing @xmath22 , will probe a less shallow potential ( panel b ) resulting in a stronger hardening of @xmath19 than the harmonic one and leading to @xmath130 .
a more complex situation is found for @xmath112 . here ,
more than a softening of the quadratic term @xmath131 , increasing @xmath18 the lattice potential @xmath119 presents two flat regions at finite @xmath63 which evolve first in two relative minima and then in two absolute minima lower than the central one ( panel c ) . in this framework
we distinguish two regimes .
the first one is when the zero point energy is roughly lower than the flat regions ( or minima ) of the potential ( @xmath132 ) ; in this situation increasing @xmath22 makes the lattice wave - function more distorted , the metallic character is reduced and the phonon energy is softened with respect to the harmonic one .
this phenomenology is thus reflected in the downturn of @xmath85 and of @xmath113 .
a second regime is reached for @xmath18 closer to the mit ( @xmath133 ) : here the minima are lower than the zero point energy , a phenomenology similar to the double - well potential is restored and @xmath134 , @xmath130 .
the results for @xmath135 ( panels e , f ) are essentially similar to those for @xmath112 for small @xmath18 . here
however the first - order metal - insulator transition occurs well before the relative minima at finite @xmath63 reach the range of the zero point energy . as a consequence
the previous considerations apply , namely @xmath85 shows an downturn ( but it still remains finite ) and @xmath113 .
the evolution of the lattice potential as function of @xmath2 and @xmath18 sheds interesting light on the origin of the first - order / second - order character of the mit .
it has been indeed claimed that both the second - order at small @xmath2 and the first - order transition at large @xmath2 could be related to a phonon - driven mechanism.@xcite in a phonon - driven ginzburg - landau theory the character of the transition is thus determined by the dependence of the free energy on the lattice coordinate , @xmath136 , which in the adiabatic limit reduces to the lattice potential @xmath119 .
the second - order transition at small @xmath2 is thus associated with the development of a double - well potential which occurs for @xmath137 and @xmath138 , whereas a first order one is expected to correspond to a three - well structure with absolute minima at @xmath139 , condition realized for @xmath140 , @xmath141 .
our results shown in fig . [ f - psvx ] confute however this picture . at @xmath142
the second order mit transition occurs _
after _ the development of the double well in @xmath119 .
mit here is indeed a by - product of the lattice distortions as was shown by adiabatic analysis of ref . and is due to the vanishing of the coherent tunneling processes between the minima of two double - well lattice potential . for @xmath107
instead the developing of the three - well shape of the lattice potential is not sufficient to explain the first - order character since the mit occurs far _ before _ the local minima with finite lattice distortions at @xmath143 become absolute minima .
moreover the energy of the local minima at the transition become higher and higher in energy as @xmath2 increases , precisely where the first - order character becomes more marked .
these considerations suggest thus that the origin of the first - order metal - insulator transition in the strongly correlated regime is likely ascribed to an electronic origin more than some phonon - driven instability .
an intrinsic difference between the weakly and the strongly correlated metallic phase is the presence of a spin - singlet character in the metallic case pointed out by the presence of the hubbard bands coexisting with the kondo peak in the strongly correlated regime . if the metallic state has a pronounced spin - singlet character the transition to a bipolaronic pair state ( which is instead a charge pseudo - spin singlet ) can be viewed as a level crossing between this two different electronic states therefore leading to a first order transition .
in the previous section we have analyzed in details the behavior of the ies on the physical quantities as function of the electron - phonon coupling @xmath18 .
we have shown in particular that the ies are quite sensitive to the different regimes of the electronic correlations , and we have distinguished the weak from the highly correlated regime .
the aim of this section is to address this issue in more details by studying the dependence of the ies on the hubbard repulsion @xmath2 .
we shall focus on the ie on the quasi - particle spectral weight @xmath0 and on the renormalized phonon frequency @xmath19 which show the most characteristic features .
we consider two values of @xmath18 , representative of two different physical regimes : @xmath144 , for which the system is far from the bipolaronic phase even in the absence of any hubbard repulsion ; and @xmath145 for which value instead the ground state is a bipolaronic insulator at @xmath146 . in this case one
needs a finite value of the hubbard repulsion ( @xmath147 ) to contrast the el - ph driven attraction and to restore the metallic character .
both the cases of course undergo a mott metal - insulator transition as @xmath2 approaches some critical value @xmath148 . in fig .
[ trend - u ] we report the evolution of the quasi - particle spectral weight @xmath0 and of the renormalized phonon frequency @xmath19 as functions of @xmath2 for the two chosen @xmath18 .
let us discuss first the behavior of @xmath19 , which shows a monotonic behavior for @xmath144 .
this trend is representative for the generic case of weak electron - phonon coupling : due to the suppression of the double / empty sites , increasing @xmath2 leads to a reduction of the charge fluctuations and to the consequent enhancement of the renormalized phonon frequency @xmath19 which results less screened . more intriguing
is the analysis of @xmath0 which , unlike the pure hubbard model , shows a non - monotonic behavior as function of @xmath2 . in the previous section
we have discussed how , in the strongly correlated regime close to the mott mit , switching on the electron - phonon interaction leads to an increase of @xmath0 .
this trend has been explained as a result of the competition between the attractive el - ph coupling and the repulsive hubbard term.@xcite fig .
[ trend - u ] shows that a similar mechanism applies also close to the bipolaron mit where the phonon - driven attraction is the strongest interaction , and the hubbard repulsion partially counteracts it , so that @xmath0 increases with the repulsion for small @xmath2 . for larger
@xmath2 however the repulsion overcomes the attraction , the system approaches the mott mit and @xmath0 recovers the standard behavior ( @xmath78 ) .
these trends are also more marked for @xmath145 where , as mentioned above , the system is a ( bipolaronic ) insulator in the absence of hubbard repulsion , and a finite @xmath149 is required to have a metallic state .
most evident is of course the competition between @xmath2 and the electron - phonon interaction , which becomes a sharp transition from @xmath150 for @xmath151 to a maximum value @xmath152 for @xmath153 . in the right panels of fig .
[ trend - u ] we show the corresponding ies @xmath85 , @xmath93 . also here the behavior of the isotope effect on the quasi - particle spectral weight is particularly interesting . in section [ s - weak ]
we have shown that giant ies on the quasi - particle spectral weight @xmath0 are expected as the system approaches the bipolaronic metal - insulator transition . from an intuitive point of view
we can expect that the electronic correlation , driving the systems far from the bipolaronic transition , reduces the magnitude of the anomalous ie @xmath85 .
this picture is confirmed in fig .
[ trend - u ] where we see that @xmath85 is strongly reduced as soon as @xmath2 moves the system away from the bipolaronic mit .
this trend is however quite soon followed by a further enhancement of the anomalous ie on @xmath0 which becomes larger as the systems approaches the mott transition , revealing that the strongly correlated regime close to the mott metal - insulator transition is also highly sensitive to the phonon dynamics .
this counterintuitive result can be qualitatively understood by following the analysis of ref . which showed that the strongly correlated regime close to the mott mit _ in the presence _ of electron - phonon interaction could be described in terms of a pure hubbard model with a properly rescaled hubbard repulsion @xmath154 , where @xmath155 .
since the quasi - particle spectral weight @xmath0 is known to scale linearly with @xmath156 in dmft for @xmath157 , where @xmath158 is the critical value of the hubbard repulsion at which a mott metal - insulator transition occurs , we obtain thus @xmath159 this expression shows that the isotope effect on @xmath0 diverges as @xmath160 as @xmath161.@xcite the analytical behavior predicted by eq .
[ analytical ] , is also shown in the right - top panel of fig .
[ trend - u ] ( dashed lines ) .
the critical values @xmath158 for each @xmath18 have been chosen in order to have the best fit with the dmft numerical results : namely @xmath162 for @xmath144 and @xmath163 for @xmath145 .
the crossover between the two different correlation regimes is also pointed out by the analysis of the behavior of the ies on the renormalized phonon frequency @xmath93 .
most evident is the case of @xmath145 where in the absence of hubbard repulsion the ground state is a bipolaronic insulator . here
the non - monotonic behavior of @xmath93 reflects two different regimes : @xmath164 ) a first sharp region @xmath165 where @xmath166 . in this regime
the system is characterized by a double - well lattice potential with a zero point energy larger than the barrier between the two minima as shown in fig .
[ f - psvx]b . by increasing @xmath22
the system will probe thus a less shallow potential resulting in a stronger hardening of @xmath19 than the harmonic one and leading to @xmath130 .
@xmath167 ) for @xmath168 , on the other hand , the shape of the lattice potential @xmath119 is more similar to the panel ( f ) of fig .
[ f - psvx ] where the zero point energy is lower than the unstable minima .
increasing @xmath22 the system would thus probe the flat potential regions around these two minima leading to an increase of the renormalized phonon frequency smaller than the harmonic one ( @xmath113 ) .
these effects become smaller as the hubbard repulsion @xmath2 increases since the minima shift higher in energy and the ie @xmath93 approaches asymptotically @xmath169 for @xmath170 .
a similar behavior occurs for @xmath144 where however the system in the absence of electronic correlation is still metallic although close to the bipolaronic mit . in this case however increasing @xmath2 leads to a reduction of the ie on @xmath19 until , for @xmath171 , @xmath93 becomes smaller than @xmath169 , although in a less marked way than in the @xmath145 case because of the smaller electron - phonon coupling .
the evidence of this anomalous ie @xmath113 even for small @xmath18 suggests thus that this regime is not related to the closeness to the bipolaronic metal - insulator transition .
it is worth to stress in addition that the evidence of these two regimes of correlation prompts out from the analysis of the ie @xmath93 while the renormalized phonon frequency itself @xmath19 does not show any signature of this crossover .
in this paper we have studied in details the isotope dependence of several physical quantities : the quasi - particle spectral weight , the renormalized phonon frequency , and the static component of local charge and spin susceptibilities in the metallic regime of the hubbard - holstein model within dmft approximation .
the work is carried out for the half - filled model in the metallic phase without broken symmetries showing anomalous behaviors approaching the pair and the mott metal - insulator transitions .
we have shown that the anomalies of isotope coefficients can be helpful in revealing different physical regimes .
in particular , we have found that in the adiabatic regime @xmath105 we can identify two different regimes according to the degree of correlation .
for @xmath172 the physical properties of the system are qualitatively similar to those of the pure holstein model where the approach to the bipolaronic metal - insulator transition obtained within dmft is reflected in a near divergence of the ies characteristic of bipolaronic binding . in this regime
we find a divergent @xmath85 and a strong increase of @xmath90 close to the pair mit while both @xmath93 and @xmath94 have a near divergence at @xmath72 ( @xmath91 and @xmath92 ) .
the second regime establishes for @xmath173 , where the bipolaronic mit becomes of the first order .
the change in the order of the transition can be characterized in an accurate way by means of our study of ies . in the highly correlated regime
, the isotope effects are indeed not diverging even when the first - order transition is approached .
the lack of any divergence as well as an analysis of the lattice potential suggest that the first - order character is more likely ascribed to an electronic mechanism than to a lattice - driven instability .
in addition we have shown that approaching the mott - hubbard mit also gives rise to drastic anomalous ies .
we acknowledge useful discussions with c. castellani , l. pietronero and g. sangiovanni .
this work was partially funded by the miur project firb rbau017s8r and prin projects 2003 and 2005 .
d. rubio temprano , j. mesot , s. janssen , k. conder , a. furrer , h. mutka , and k. a. mller , phys .
lett . * 84 * , 1990 ( 2000 ) ; d. rubio temprano , j. mesot , s. janssen , k. conder , a. furrer , a. sokolov , v. trounov , s. .
kazakov , j. karpinski , and k. a. mller , eur .
j. b * 19 * , 5 ( 2001 ) . | we study the isotope effects arising from the coupling of correlated electrons with dispersionless phonons by considering the hubbard - holstein model at half - filling within the dynamical mean - field theory .
in particular we calculate the isotope effects on the quasi - particle spectral weight @xmath0 , the renormalized phonon frequency , and the static charge and spin susceptibilities . in the weakly correlated regime @xmath1 , where @xmath2 is the hubbard repulsion and @xmath3 is the bare electron half - bandwidth , the physical properties are qualitatively similar to those characterizing the holstein model in the absence of coulomb repulsion , where the bipolaronic binding takes place at large electron - phonon coupling , and it reflects in divergent isotope responses . on the contrary in the strongly correlated regime @xmath4 , where the bipolaronic metal - insulator transition becomes of first order , the isotope effects are bounded , suggesting that the first order transition is likely driven by an electronic mechanism , rather then by a lattice instability .
these results point out how the isotope responses are extremely sensitive to phase boundaries and they may be used to characterize the competition between the electron - phonon coupling and the hubbard repulsion . |
the study of eclipsing binary star systems ( hereafter ebs ) has a long history .
the eclipse hypothesis was proposed as an explanation of the periodic dimmings of the ` demon star ' algol ( @xmath0persei ) by john @xcite .
the first empirical measurement of the masses and radii of two stars in an eb was that by @xcite for @xmath0aurigae ; the numbers are close to modern values @xcite . shortly afterwards , @xcite established a quantitative method for analysing light curves of eclipses .
the era of high - quality space photometry began in 2006 , with a light curve of @xmath1centauri from the wire satellite @xcite .
this star shows deep total eclipses on a 38.8d period ( fig.[fig : pcen ] ) , and prior to the serendipitous observations from wire was not known as either an eclipsing or spectroscopic binary despite its brightness ( @xmath2 ) .
the current era of vast photometric surveys has led to the discovery of thousands of new ebs , with extensive light curves being obtained particularly by surveys for transiting planets ( e.g. tres , hat , wasp , corot , _ kepler _ , and in the future tess and plato ) .
fig.[fig : comp ] compares the light curves of a transiting planetary system and eb : it is clear that any decent photometric survey for transiting planets is easily capable of detecting eclipsing binaries .
centauri @xcite.,scaledwidth=80.0% ] the importance of ebs lies in their amenability to detailed analysis . from fitting a light curve with a simple geometrical model
one can determine the fractional radii of the two stars ( @xmath3 and @xmath4 where @xmath5 are the true radii of the stars and @xmath6 is the orbital inclination of the system ) .
spectroscopic radial velocity measurements for both stars allow the minimum masses ( @xmath7 and @xmath8 ) and orbital separation ( @xmath9 ) to be calculated .
these quantities immediately yield measurements of the masses and radii of the two stars , which can be achieved empirically and to accuracies better than 1% if the observational data are of good quality ( e.g. * ? ? ?
obtaining the @xmath10 values of the stars procedures for which usually incur some dependence on theoretical models
leads immediately to determination of their luminosities using the formula @xmath11 , making ebs excellent distance indicators .
eb - based distances are available for open clusters as well as for nearby galaxies ( e.g. * ? ? ?
* ; * ? ? ?
another use of ebs is as tests and calibrators of stellar evolutionary theory .
the predictions of theoretical models are required to match the measured values of mass , radius , @xmath10 and luminosity for two stars of the same age and chemical composition ( e.g. * ? ? ?
* ; * ? ? ?
many ebs contain stars which show intrinsic variability due to pulsations .
stochastic oscillations are observed in solar - type dwarfs , are present with much higher amplitudes in cool giant stars , and can be used to determine the density , radius and thus mass of the star to high precision @xcite .
although of particular importance for solar - type dwarfs , stochastic oscillations have so far only been detected in the giant components of ebs .
other types of pulsations which have been found in conjunction with eclipses are @xmath12scuti , @xmath13doradus , spb and tidally - induced oscillations . in this work
i review recent results for these objects , then consider the future impact of current and forthcoming space missions .
the one type of system where properties from asteroseismology can currently be compared to those determined via the effects of binarity is that of eclipsing red giants .
a single eclipse , 0.12mag deep and lasting 2.2d , was found in the quarter 1 light curve of kic8410637 @xcite obtained by the _ kepler
_ satellite .
subsequent spectroscopic follow - up observations coupled with further eclipses found in the _ kepler _ data have yielded accurate mass and radius measurements for the giant ( @xmath14@xmath15 , @xmath16@xmath17 ) and its companion ( @xmath18@xmath15 and @xmath19@xmath17 ) , which pursue orbits of period 408.3d and eccentricity @xmath20 @xcite .
the properties of the giant from asteroseismology ( @xmath21@xmath15 , @xmath22@xmath17 ; @xcite ) are larger than the results from the eb analysis by @xmath23 and @xmath24 , respectively , for reasons which are currently under investigation .
@xcite has found 12 more similar systems using _ kepler _
data , with orbital periods ranging from 14.7d to 1058d . @xcite noticed that the giants in the shorter - period ebs do not show stochastic oscillations , to a detection limit far below the oscillation amplitudes expected for such objects .
this suggests that tidal effects spin up the giant , causing magnetic fields which dissipate the acoustic modes .
@xcite found that mode depletion occurred when the fractional radius of the giant was larger than 0.160.24 .
hd181068 ( kic5952403 ) is a bright giant observed by _
it is the primary component of a hierarchical triple system and is orbited by a short - period binary containing two k - dwarfs @xcite .
the two dwarfs are eclipsing on a 0.9d period , and themselves transit and are occulted by the g - type giant on a 45.5d period .
the giant star shows no stochastic oscillations such as are found in single stars of similar mass and radius , but instead shows oscillations at multiples of the orbital frequency .
@xcite found four frequencies in the _ kepler _ quarter 111 data : all are related to the orbital frequency and at least two are tidal in origin .
@xcite determined the full physical properties of this triple system by using eclipse timing variations in the short - period eb as a substitute for rvs of the two dwarfs , which are unavailable due to their relative faintness .
tidally - induced pulsations have previously been seen in hd174884 ( corot7758 ) , an eb consisting of two unevolved b - stars @xcite .
scuti eb kic10661783 @xcite .
top : part of the short - cadence observations . bottom : phased light curve after removal of the pulsations.,title="fig:",scaledwidth=80.0% ] scuti eb kic10661783 @xcite .
top : part of the short - cadence observations .
bottom : phased light curve after removal of the pulsations.,title="fig:",scaledwidth=80.0% ] @xmath12scuti stars are a- and f - type dwarfs which pulsate in non - radial @xmath25-modes with periods from 0.014d to 0.333d @xcite
. a substantial number of known @xmath12scuti stars are members of ebs , including both detached and semi - detached systems .
@xcite defined the class of @xmath12scuti stars in semi - detached systems , which have since been labelled as ` oea ' ( oscillating algol ) stars .
recent advances in the asteroseismic analysis of @xmath12scuti stars have enabled the detection of large frequency separations analogous to solar - like oscillations but in a different pulsation regime @xcite , representing a step towards utilising these pulsators as tests of stellar physics . @xcite
have used ebs observed with _ kepler _ and corot to define a relation between the large frequency separation and mean density for @xmath12scuti stars . an early result from the _ kepler
_ satellite was the detection of eclipses and 68 pulsation frequencies in the light curve of kic10661783 @xcite .
follow - up spectroscopy by @xcite allowed determination of the masses and radii of the two stars .
the secondary component appears to be a classic example of close - binary evolution , having a mass of @xmath26@xmath15 and a radius of @xmath27@xmath17 .
however , the spectroscopic mass ratio leads to a light curve model in which the two stars are currently detached and therefore not undergoing mass transfer .
kic10661783 may be an example of a semi - detached system which shows only episodic mass transfer .
other examples include kic3858884 @xcite , a detached eb with an orbital period of 26.0d and a high eccentricity ( @xmath28 ) .
theoretical models require convective core overshooting to match the properties of this system .
corot 105906206 is another , for which @xcite required the inclusion of doppler beaming to fit the light curve .
dor pulsations.,scaledwidth=80.0% ] scuti and @xmath13dor pulsations .
top : part of the short - cadence observations . bottom : close - up of an short stretch of data outside eclipse.,title="fig:",scaledwidth=80.0% ] scuti and @xmath13dor pulsations .
top : part of the short - cadence observations .
bottom : close - up of an short stretch of data outside eclipse.,title="fig:",scaledwidth=80.0% ] @xmath13dor stars are non - radial @xmath29-mode pulsators with typical periods of 0.4d to 3d @xcite and spectral types of a7-f5 v - iv .
the advent of high - precision photometry from space satellites has revealed that a large fraction are hybrids which show both @xmath13dor and @xmath12scuti pulsations @xcite .
recent advances in asteroseismology have enabled the detection of period spacings which hold the prospect of detailed seismic analyses of these objects @xcite .
it was inevitable that some @xmath13dor stars would be found in ebs , and cases have indeed been identified from _ kepler _ observations .
@xcite presented a detailed analysis of kic11285625 , which shows 12%-deep eclipses on a 10.8d period plus @xmath29-mode oscillations of similar amplitude ( fig.[fig:1128 ] ) .
analysis of the _ kepler _ data and ground - based radial velocity measurements allowed debosscher et al . to determine the masses and radii of the two component of the eb to 1% .
amplitude modulation and splitting at the rotational and orbital frequencies was also found from analysis of the pulsations .
another example is kic4544587 @xcite , an eb with an eccentric 2.2d period orbit and rapid apsidal motion . whilst the eclipses
dominate the _ kepler _ light curve ( fig.[fig:0454 ] ) , both @xmath12scuti and @xmath13dor pulsations are visible on closer inspection . @xcite found 14 @xmath29-modes , most of which were harmonics of the orbital frequency , and 17 @xmath25-modes which were either affected by tides or occurring within the secondary star .
a major goal for space - based transit surveys such as _ kepler _ and plato is the identification of potentially habitable planets .
the widely accepted criteria for such planets is that they are earth - like : rocky and orbiting within the habitable zone of their host star ( e.g. * ? ? ?
the low masses and radii of such planets , coupled with their long orbital periods , makes them difficult to characterise .
their radii will be measured relative to those of their host stars , so it is important to obtain as much information as possible about these stars .
asteroseismology is an important component of such work ( e.g. * ? ? ?
* ; * ? ? ?
* ) , and ebs offer an unrivalled opportunity to check the reliability of the asteroseismic results
. however , this has yet to be achieved because stochastic oscillations have not been detected in a solar - type eclipsing binary .
the short - period ebs observed using _
suffer from tidal effects which cause rotational splitting and dissipation of the acoustic modes , making them undetectable in current data .
the scarcity of longer - period ebs ( orbital period greater than approximately 10d depending on the masses and evolutionary statuses of the components ) means that no suitable candidates have been found for a comparison between asteroseismic properties and those measured from eclipses and radial velocities .
one notable success has been the detection of stochastic oscillations in the extremely bright ( @xmath30 ) totally - eclipsing binary v380cygni @xcite , which is composed of a b1.5 giant and a b2 dwarf of known chemical composition @xcite .
fig.[fig : v380 ] shows part of the _ kepler _ light curve of this system , in which can be seen primary and secondary eclipses , periastron brightening due to tidal effects , and stochastic variability .
a fourier transform of the _ kepler _ light curve after subtraction of the effects of binarity shows some frequencies which are multiples of the orbital frequency and others which are not .
the physical properties of the binary can only be matched by theoretical models with unrealistically high rotation and strong convective core overshooting , implying that much increased mixing is needed in these models .
although this review has showcased many results based on data from the _ kepler _ satellite , the archive of _ kepler _ data contains over 2700 ebs ( kirk et al .
, in press ) of which the vast majority have not received significant attention .
the corot satellite also obtained over 163000 light curves including over 3000 ebs @xcite , and the k2 mission is continuing to accumulate copious high - quality space - based data @xcite .
the near future holds the prospect of the tess satellite , which will observe most of the sky for 27d intervals and thus greatly expand the database for bright ebs .
one of the remaining tasks is to study solar - type ebs for which stochastic oscillations can be detected .
this necessarily requires space - based photometry in order to get observations with a high duty cycle , low scatter , and long time coverage .
each of the past and present space missions has at least one feature which compromises its ability to perform this experiment , such as low sampling rate ( most of the _ kepler _ targets ) , faintness of the stars observed ( _ kepler _ and corot ) , or insufficient duration of observations ( tess , except for small patches of sky near the celestial poles ) . the plato mission promises to solve all of these problems @xcite .
scheduled for launch in 2024 , it will observe hundreds of thousands of bright stars at high cadence and for time intervals up to several years .
extensive effort will be needed to cope with this inundation of data , and to procure follow - up observations from ground - based facilities .
the work package structure for plato includes one dedicated to binary and multiple stars , for which the author is responsible , within the complementary science work package group .
although preparation of an extensive target list of known ebs will form part of this work package , it is likely that the majority of ebs observed by plato will be previously unidentified binaries selected for observation in one of the main transit search samples .
these will , by design , be f , g , k and m dwarfs and the study of those showing eclipses promises to revolutionise our understanding of solar - type and low - mass stars .
, d. e. , kusakin , a. v. , gamarova , a. y. , nazarenko , v. , 2002 , in c. aerts , t. r. bedding , & j. christensen - dalsgaard , ed . , iau colloq . 185 : radial and nonradial pulsations as probes of stellar physics , vol . 259 of _ astronomical society of the pacific conference series _ , p. 96 | the study of eclipsing binaries is our primary source of measured properties of normal stars , achieved through analysis of light and radial velocity curves of eclipsing systems .
the study of oscillations and pulsations is increasingly vital for determining the properties of single stars , and investigating the physical phenomena active in their interiors . combining the two methods
holds the promise of establishing stringent tests of stellar evolutionary theory , and of calibrating model - dependent asteroseismology with empirically measured stellar properties .
i review recent advances and outline future work . |
deep - inelastic scattering ( dis ) provides a wealth of information about nucleon structure . recently ,
very high momentum transfers in dis have been achieved at the hera collider , where 820 gev protons have been collided with 27.5 gev positrons for a center - of - mass energy @xmath4=300 gev . in the highest momentum transfer region , @xmath5 dis cross sections depend on proton parton densities and properties of the electroweak interaction .
the @xmath6 dis process is illustrated in fig .
[ fig : dis ] . the variables used to describe the process are @xmath7 , the struck parton momentum fraction , @xmath8 , the fractional energy transfer in the proton rest frame ( inelasticity ) and @xmath0 , the four - momentum transfer squared , where @xmath9 .
neutral - current dis events are characterized by the exchange of a photon or @xmath10-boson , and have a positron and a jet ( or jets ) in the final state .
the outgoing positron and the hadronic matter are balanced in transverse momentum . charged - current dis events are characterized by the exchange of a @xmath11-boson , and contain an undetected neutrino and a jet ( or jets ) in the final state .
the presence of the neutrino is detected as missing transverse momentum .
the dis data presented here were collected and analyzed by the zeus collaboration and correspond to an integrated luminosity of 47.7 pb@xmath1 taken from 1994 to 1997 .
zeus @xcite is a multi - purpose magnetic detector ; the primary components used in these analyses are the calorimeters ( rcal , bcal , fcal ) , the central tracking detector ( ctd ) , and the luminosity monitor .
the coordinate system is defined such that the @xmath12-axis follows the proton direction , and the origin is the nominal @xmath13 interaction point . the zeus detector is displayed in fig . [
fig : detector ] .
the zeus compensating uranium - scintillator calorimeter covers the polar angle region @xmath14 with full azimuthal coverage over this region .
its energy resolution for electromagnetic showers is @xmath15 , and for hadronic showers is @xmath16 , as measured under test - beam conditions .
the zeus ctd operates in a solenoidal 1.43 t magnetic field , and primarily provides vertex reconstruction , track momentum , and charge information for these analyses .
the luminosity is determined from the rate of bethe - heitler bremsstrahlung ( @xmath17 ) photons detected in an electromagnetic calorimeter at @xmath18 m. 13 truecm
to lowest order ( qed born level ) the neutral - current dis ( @xmath19 ) cross section is @xmath20\ ] ] where @xmath21 . in lowest - order qcd ,
the structure functions @xmath22 and @xmath23 are the sums over quark flavor of the product of quark couplings and momentum distributions .
the quark couplings depend on the quark charges , and the electroweak parameters @xmath24 , etc .
the qed born - level charged - current dis ( @xmath25 ) cross section is @xmath26\ ] ] where at lowest order the structure functions @xmath27 and @xmath28 contain sums and differences of quark and antiquark momentum distributions .
the neutral- and charged - current longitudinal structure functions , @xmath29 and @xmath30 , respectively , provide a small ( @xmath31 ) contribution in the kinematic range discussed here , and have been included .
electroweak radiative corrections to these born - level equations , including initial- and final - state radiation , vertex and propagator corrections , and two - boson exchange , are significant and have been included to at least lowest order @xcite .
the differential nc dis cross section @xmath32 is shown in fig . [
fig : nc_dsdq2 ] ; @xmath33 and @xmath34 are shown in fig . [
fig : nc_dsdx_dsdy ] .
the differential cc dis cross sections @xmath35 ,
@xmath36 , and @xmath37 , are shown in figs .
[ fig : cc_dsdq2 ] , [ fig : cc_dsdx ] . and [ fig : cc_dsdy ] ,
respectively . for both neutral and charged current , the data points are compared to the standard model predictions using the cteq4d @xcite parton distribution functions ( pdf s ) , shown by the solid curves , with estimated pdf uncertainty shown by the shaded bands .
5.1 truecm 5.1 truecm 9.7 truecm 9.7 truecm 6.7 truecm 6.7 truecm 6.4 truecm 6.4 truecm 6.4 truecm 6.4 truecm the pdf uncertainty is calculated from a nlo fit @xcite to world dis data , and includes statistical and systematic errors on these data , as well as variations in the assumed electroweak and qcd parameters . for the neutral - current cross section
, these uncertainties range from 2.5% at @xmath38 to 8% at @xmath39 . for the charged - current cross section ,
the extracted uncertainties range from 9% at @xmath38 to 17% at @xmath40 .
the larger cc uncertainty is due to the larger uncertainty in the @xmath41-quark pdf relative to the @xmath42-quark pdf .
both the nlo fit , which includes higher - twist effects , and a recent reanalysis of nmc and slac data @xcite yield a larger @xmath43 ratio at high-@xmath7 than the cteq4 pdf s , where the @xmath43 ratio is constrained to be zero at @xmath44 . within present experimental precision ,
any of these hypotheses can be accommodated ; their differences are not included in the pdf uncertainty band .
increasing the @xmath43 ratio at high-@xmath7 reduces the cc data excess at high-@xmath7 , but does not appreciably affect the nc cross section since the nc process is not sensitive to the @xmath41-quark . the charged - current dis reduced cross section @xmath45
is shown in fig .
[ fig : ccreduced ] along with the standard model ( cteq4d ) prediction . at high-@xmath7 , the valence @xmath41 and @xmath46 quarks ( dashed curves ) dominate @xmath47 , whereas at lower-@xmath7 the @xmath48 and @xmath49 sea quarks ( dotted curves ) dominate
. 12 truecm at momentum - transfer - squared close to the @xmath10 and @xmath11 masses squared , i.e. , @xmath50 gev@xmath3 , the cross sections become sensitive to contributions from these exchanges , and to the propagator masses in particular .
the sensitivity of the nc cross section to the @xmath10 mass is demonstrated in fig .
[ fig : zmass ] .
the measured cross sections are compared with the standard model predictions by varying @xmath51 while keeping the couplings fixed .
three mass values are considered , @xmath51 = 40 , 91 and @xmath52 gev . clearly , @xmath53 gev is favored , in agreement with the world average value @xmath51 = ( 91.187 @xmath54 0.007 ) gev @xcite . in the cc channel , the sensitivity of the cross section to the @xmath11 mass is better than in the nc case , as shown in fig .
[ fig : wmass ] . the @xmath55 may be extracted by a @xmath56 fit to the measured cross sections , leaving all other electroweak parameters fixed , with the result @xmath57 this may be compared with the world average value of @xmath55 = ( 80.41 @xmath54 0.10 ) gev @xcite . the agreement between these @xmath10 and @xmath11 masses and the masses extracted through timelike production in @xmath58 and @xmath59 data confirms the standard model prediction in the spacelike regime . 5.0 truecm 5.0 truecm 10 truecm
the zeus neutral - current and charged - current dis cross sections in the momentum transfer range @xmath60 gev@xmath3 , i.e. , for @xmath0 over more than two orders of magnitude , have been shown to be in good agreement with the standard model predictions . however
, a slight excess persists at highest @xmath0 . with the current data set , zeus has gained sensitivity to the electroweak propagator masses @xmath55 and @xmath51 .
in 1998 , hera switched from positron - proton collisions to electron - proton collisions .
in addition , the proton energy was increased from 820 gev to 920 gev , thus increasing the center - of - mass energy by 6% from 300 gev to 318 gev . as of january 1 , 1999
, zeus had already accumulated @xmath61 of @xmath62 data , nearly an order of magnitude more than accumulated during the only other @xmath62 running period 1992 - 1993 .
an additional @xmath63 is expected in the 1999 calendar year .
the hera upgrade begins in may 2000 , after which we foresee @xmath64/year .
these future data will improve our understanding of the electroweak interaction , and may even hold some surprises .
we anticipate an exciting future for high-@xmath0 physics at hera .
the kind assistance of my zeus colleagues in the preparation of this talk , and the great efforts of the conference organizers are greatly appreciated .
this work is partly funded by the u.s .
department of energy .
zeus collaboration , the zeus detector status report , desy 1993 .
a. kwiatkowski , h. spiesberger , and h .- j .
mhring , comp .
* 69 * ( 1992 ) 155 ; + a. arbuzov et al . ,
* 94 * ( 1996 ) 128 .
lai , et al .
d * 55 * , 1280 , ( , 1997 ) .
m. botje , desy 99 - 038 , nikhef 99 - 011 ( in preparation )
yang and a. bodek , phys .
lett . 82 ( 1999 ) 2467 .
c. caso et al . , eur .
j. * c3 * ( 1998 ) 1 . | neutral - current and charged - current deep - inelastic scattering at very high four - momentum transfer squared ( @xmath0 ) have been studied in positron - proton collisions at center - of - mass energy 300 gev using the zeus detector at hera .
an integrated luminosity of 47.7 pb@xmath1 was collected in the years 1994 - 1997 .
differential cross sections are presented for @xmath2 gev@xmath3 and compared to standard model predictions . |
the biggest , and oldest , unresolved enigma in fundamental particle physics is : _ where do the observed masses of elementary particles come from ? _ , the concept of mass is not really understood , and their numerical values remain a mystery . there is the widespread , but erroneous , belief that when the higgs boson is confirmed , the origin of mass has been found .
this is not the case .
it merely replaces one set of unknown parameters ( particle masses ) with an equally unknown set of parameters ( coupling constants to the higgs field(s ) ) , so nothing is gained in the _ fundamental understanding _ of masses . despite what can be imagined about the triviality regarding masses from frequent statements like
the supersymmetric partners differ _ only _ in their masses ( and spin ) compared to the normal particles " , ... the planck scale , @xmath1 gev " , and the mass of the top quark is 175 gev " , etc , the concept of mass has never been defined in an unambiguous way , not even in classical physics .
in fact , jammer @xcite has been able to write two whole monographs on this thorny subject , concluding that nobody knows what mass really is .
furthermore , and this is a problem even for the most pragmatic of physicists , the mass parameters experimentally measured for elementary particles have no theoretical explanation whatsoever . from the vantage point of theory the masses
could just as well be a set of randomly generated numbers . to quote richard feynman : + ...
although people say that there are no experiments to lead us , it s not true .
we have some twenty - four or more - i do nt know the exact number - mysterious numbers associated with masses .
why is it that the mass of the muon compared to that with the electron is exactly 206 or whatever it is , why are the masses of the various particles such as quarks what they are ? all these numbers , and others analogous to that - which amount to some two dozen - have _ no _ explanations in these string theories - absolutely none !
there s not an idea at the present time , in any of the theoretical structures that i have heard of , which will give a clue as to why those masses are what they are ... when you look at these numbers , they look absolutely random and hectic ; there does nt seem to be much pattern in them . that s a problem for theoretical physics , and these superstring theories do nt address it at all . " @xcite
+ throughout this entire story there remains one especially unsatisfactory feature : the observed masses of the particles , @xmath2 .
there is no theory that adequately explains these numbers .
we use the numbers in all our theories , but we do nt understand them - what they are , or where they come from . i believe that from a fundamental point of view , this is a very interesting and serious problem . "
@xcite , p.152 .
the fundamental concept of mass is still so poorly understood that it was considered worthwhile to test whether an anti - atom ( produced at cern for the first time @xcite ) falls up or down in a gravitational field @xcite .
mass was smuggled " into the framework of quantum mechanics , being the same parameter as in classical physics , _
i.e. _ , the proportionality , or inertial , constant between the acceleration of a body and the applied force . however , as neither force nor acceleration are appropriate concepts in quantum mechanics , except through ehrenfest s theorem @xcite , as statistical or ensemble " entities , this direct translation of the concept is somewhat dubious . in quantum mechanics and
quantum field theory the concept of mass as resistance against acceleration " is simply inappropriate .
the most legitimate definition of mass in quantum mechanics seems to be by the relation @xmath3 , or more correctly @xmath4 @xmath5 , _
i.e. _ , as something empirically measurable , _ e.g. _ , by annihilation / production of particle - antiparticle pairs . as all other _ observables
_ in quantum mechanics , mass should be represented by an _ operator _ with its resulting smeared out probability distribution ( unless an exact eigenstate ) , peaked at the classical value . naively , the only way to get discrete values for the mass would be to assume that the elementary " particles are bound states of something more fundamental @xcite , @xcite . probably the only way to end this infinite regress to smaller scales ,
would be to assume that the ultimate subconstituents are massless .
the mass would then necessarily have to arise strictly from the dynamics , mass without mass " , john wheeler s goal to remove mass from the _ basic _ equations of physics @xcite - as implicitly anticipated by einstein in his article _ does the inertia of a body depend upon its energy content ? _ : the mass of a body is a measure of its energy - content " @xcite . according to mach s principle " @xcite the mass of a particle arises because of its interaction with the rest of the universe .
this would mean that mass would no longer be a scalar but a tensor ( of second rank ) , as inhomogeneities and anisotropies in the distribution of surrounding matter would give a directional and temporal dependence in the inertial mass .
this is accidentally similar to the modern parametrization of mass as a matrix ( tensor of second rank ) when discussing fermion masses @xcite in general , and neutrino oscillations in particular . as none of the speculative theoretical `` advances '' since the 1970s , such as grand unified theories ( guts ) , supersymmetry
( susy ) , superstrings and m - theory to name but a few , has yet yielded a single experimentally tested result , we instead ask if the mass spectrum can be understood ( and derived ) from what is known to be more or less correct , _
i.e. _ the standard model of particle physics .
although some proponents of more `` fundamental '' theories claim to , eventually , be able to deduce everything from first principles , including the mass spectrum of elementary particles , there has not yet been a single instant where this has been achieved in practice , and we thus prefer a more `` pragmatic '' approach that can be initiated _ today _ instead of waiting for the ( perhaps nonexistent ) `` ultimate '' theory to tell us all truths .
can we understand the different particle masses from what we already know , more or less , to be correct , _
i.e. _ without any `` exotic '' physics ?
we believe that this may be possible , and will elaborate on that in what follows .
we propose that the mass of a particle has a strictly local origin ( not a global one like _
e.g. _ in the assumption of mach ) , arising from its self - interaction(s ) .
that is , the mass is equivalent to the energy contained in the associated gauge fields ( in perturbative quantum field theory ; the energy of the cloud " of virtual gauge particles ) .
such a connection between fundamental dynamical interactions and mass seems only reasonable as only mass is needed to go from kinematics to dynamics @xcite . as most of the ( at least ) eighteen arbitrary parameters of the standard model " arise because of the mass problem , a connection between the masses and the ordinary , non - yukawa _ i.e. _ non - higgs , interaction couplings would also significantly reduce the number of free _ ad hoc _ parameters .
in the conventional approach to particle masses within the standard model ( and beyond ) , the higgs field generates the different masses through its coupling strength to the other fields .
these couplings are arbitrary parameters , chosen to coincide with experimental data .
so even if the higgs model turns out to be the correct one to break the electroweak symmetry ( which theoretically seems somewhat improbable as the original higgs then would be the only spinless fundamental field , with all the theoretical problems that entails ) nothing is gained in the knowledge of why the elementary particles have the masses we observe .
the higgs mechanism simply replaces one _ ad hoc _ mass parameter ( @xmath6 ) with another equally _ ad hoc _ yukawa coupling constant ( @xmath7 ) to the higgs field @xmath8 and because of this we are free to make other hypotheses as to what _ physically _ generates mass .
a higgs - like symmetry breaking mechanism is the favorite way to break also other ( hypothetical ) symmetries , such as supersymmetry . to us
that seems to be the wrong way to proceed , as such a symmetry - breaking mechanism always introduces new free ( _ ad hoc _ ) parameters , and a more fundamental theory should contain _ fewer _ free parameters , not more . as previously stated , the original standard model itself contains 18 free parameters , and several more if neutrinos are non - massless , most of them related to the theoretically incalculable masses . in supersymmetric extensions of the standard model
the free parameters rank in the number of hundreds , or more , again mainly connected to new ( unobserved ) masses of supersymmetric partners , and to new higgs - like mechanisms at higher ( incalculable ) energy scales . here
we , instead , start by assuming that _ all _ of a particle s mass arises exclusively from its interaction with itself ( see fig.1 ) .
this makes it possible for the underlying `` bare '' lagrangian to include only massless fields , just like in the higgs model before the symmetry is broken , preserving the gauge - invariance of the standard model , as the self - coupling is a vacuum phenomenon , just like the non - vanishing vacuum expectation value of the higgs field when the symmetry is broken at low " energies .
mass is thus a frozen " irreducible energy connected to , and defining , the particle . quantum field theory contribution to the mass of a particle due to self - interaction .
each vertex contributes one charge factor @xmath9 , so @xmath10 ; the mass is proportional to the _ physical _ coupling constant . ] in a _ perturbative _ quantum field - theoretic sense , the diagrams for self - interaction are of course divergent due to virtual loops , but surely nature herself is not singular .
it seems obvious that the formally infinite masses predicted by perturbative quantum field theory is only an artifact of the perturbative approximation .
( even the conventional solution " , the renormalization of masses and charges , is defined through the approximate , perturbative formulation of the theory . ) furthermore , as all gauge fields , _
i.e. _ different interactions , in the standard model diverge in the same way _ quotients _ between masses are finite .
there seems to be little reason to work at some very high ( gut / string ) scale , as the masses we are interested in are observed at `` normal '' energies .
this also spares us of evolving the behaviour at extreme fundamental " energies down to experimentally accessible energies via the renormalization group .
the renormalized , _
i.e. _ , physically measurable , mass arising from the altered propagator due to higher order perturbative corrections in quantum field theory can be written @xmath11 where @xmath12 is the `` bare '' ( non - observable ) mass , and @xmath13 ultraviolet " diverges as the spatial cut - off @xmath14 .
our approach is somewhat like taking the limit @xmath15 and @xmath16 simultaneously .
the first obvious conclusion is that the stronger the self - interaction , the more massive the particle will be .
the neutrino , interacting only through the weak force , will become very light .
next , the electron with electromagnetic coupling , becomes more massive .
the quarks , interacting also through the strong force , become heavy .
to illustrate the point more clearly , we construct a simple model .
the detailed non - perturbative dynamics will most certainly alter the ( perturbatively formally infinite ) proportionality constant for the different interactions ( due to different gauge groups ) , which in the ratio of particle masses hence will not identically cancel , but give a remaining finite factor .
for now , to obtain first approximations , we disregard this complication and assume it to be of order unity . for the mass of a particle ( to order @xmath17 )
we thus make the ansatz @xmath18 where @xmath19 is the sommerfeld fine structure constant ( @xmath20 ) and @xmath21 and @xmath22 are the ( low - energy ) couplings of the weak- , and strong interactions . as @xmath21 is ambiguous ( mass dependent )
, we infer the effective @xmath23 s from the measured characteristic reaction times ( _ e.g. _ , decay probabilities ) , as the interaction strength is but a measure of the probability of the interaction to take place .
@xmath24 gives the particle s electrical charge in units of @xmath25 , and @xmath26 and @xmath27 are analogous quantities ( of order one ) coming from the gauge - groups for `` weak charge '' and color charge " .
@xmath28 is a normalizing constant , which in a truly non - perturbative treatment of the standard model should be calculable .
however , using quotients , it cancels .
this gives the results @xmath29 @xmath30 or in mass units @xmath31 clearly compatible with direct experiments ; @xmath32 ev @xcite .
right - handed ( sterile ) neutrinos , if they exist , would be strictly massless in this scheme as they have no self - interaction . @xmath33
compatible with current quark masses @xcite .
if couplings of higher order are taken into account , the electromagnetic and weak contributions will essentially be unchanged , as they give rapidly diminishing terms in the perturbative expansion . the higher orders in @xmath22 , however ,
give a perturbatively divergent result , and hence an infinite " quark mass , in accordance with some `` explanations '' of quark confinement @xcite .
if quarks are permanently confined , it is troublesome to even define a mass for them , as wigner s mass - spin classification of elementary particles @xcite ( unitary representations of the poincar group ) is valid strictly only for free particles ( asymptotic states ) .
r.p . feynman : this repetition of particles with the same properties but heavier masses is a complete mystery . " @xcite , p.145 .
the three different particle `` generations '' of the standard model can be accounted for ( if not explained ) by a straightforward generalization of the formalism , @xmath34 where @xmath35 denotes the normalizing constants for the three different generations , @xmath36 .
if one introduces a `` generation charge '' or quantum number , @xmath37 , the @xmath35 could be written @xmath38 , where @xmath39 is a function of @xmath37 .
this gives the results ( taking the readily measured @xmath40 and @xmath41 as known ) : @xmath42 ev , @xmath43 ev , well below , and hence consistent with , the direct experimental upper limits of @xmath44 kev and @xmath45 mev @xcite . a physical mechanism for connecting the different @xmath35 s would be highly desirable , see section 5 below .
still , the number of arbitrary parameters has decreased compared to the orthodox , `` masses - from - higgs''-mechanism @xcite due to the relation between coupling strengths and masses , which in the standard model are completely independent quantities .
if we enlarge the model to include a tentative fourth generation , and use the most simple power - law ansatz , @xmath46 , utilizing the known values for the three charged leptons corresponding to @xmath47 to make a curve - fit resulting in @xmath48 , we get a prediction for the mass of an additional charged lepton @xmath49 @xcite , with @xmath50 , as @xmath51 gev , which seems ruled out as the experimental limit for a heavy charged lepton is @xmath52 gev @xcite . so a more elaborate form for @xmath39 should be sought .
however , as this would provide no _ real _ fundamental understanding , merely a parametrization , we refrain from pursuing this avenue further in this article .
an even simpler picture can be deduced by pure dimensional analysis .
if we assume that mass is directly , and solely , related to the couplings , and that the standard model is essentially `` correct '' , we only have three couplings to utilize .
these are the fine structure constant , @xmath19 , the strong interaction `` constant '' , @xmath22 , and the fermi constant , @xmath53 . to get a mass scale
, we have no choice but to use @xmath54 gev , as this has the dimension of mass ( when @xmath55 ) whereas the other two are dimensionless .
the formula for mass then takes the form @xmath56 where @xmath57 is a dimensionless function of only the fine structure constant and the strong coupling .
as the top quark mass @xmath58 is of the same order of magnitude as @xmath59 this could mean that no more massive quarks exist and , if the general structure is a true facet of nature and not just an artifact of the standard model , only three particle generations exist .
seeing that even our first - order model for the masses of elementary particles gives results of the right order of magnitude , we are now ready to discuss something more sophisticated .
truly nonlinear effects have so far received little attention in particle physics , as the dominating and inherently _ perturbative _ feynman diagram techniques mask the nonlinearities and tend to give behavior that closely mimics the linear case ( although interactions are perturbatively , _
i.e. _ mildly nonlinear ) .
we believe that the utilization of nonlinear methods , as developed in other fields of science , could greatly benefit elementary particle physics , especially in the case of masses .
the most typical feature of nonlinear equations is multiplicity of solutions , raising a hope that different generations may be automatic . in the contemporary understanding of the standard model these repetitions of particles
is a complete mystery .
interacting quantum field theories are really inherently non - linear , as can easily be seen by an analogy with classical waves .
two waves meeting in a perfectly linear medium simply penetrates each other unaffected ( think two laser beams in vacuum , or @xmath60 air ) .
this is the equivalent of non - interacting quantum field theory .
a necessary condition for the waves to interact ( _ e.g. _ , scatter ) is that the medium is non - linear .
we could thus use a classical analogy to guide our thinking in constructing a phenomenological model .
this would mean to view the vacuum as a non - linear reactive medium .
particles with different charges would thus see " a different ( effective ) medium and react differently , generating different masses . in more mathematical terms
, this can within the standard model be described by the two coupled evolution equations for gauge force fields " , @xmath61 and `` matter fields '' , @xmath62 .
@xmath63 @xmath64 where @xmath65 is the covariant derivative , @xmath66 the coupling constant ( the nonlinearity parameter " , essentially @xmath67 ) and @xmath68 the gauge field potentials .
the gauge field strength tensor is given by @xmath69 = \partial_{\mu } a_{\nu } - \partial_{\nu } a_{\mu } -i g [ a_{\mu } , a_{\nu } ] .\ ] ] we see that , even for an abelian theory like qed , where the commutators @xmath70 $ ] vanish ( physically reflecting that the photon has no self - interaction ) , the system of equations ( 10 ) , ( 11 ) , constitute a nonlinearly coupled system . in fact , in eq.(11 ) we would like to take @xmath71 ( or rather @xmath72 infinitesimal due to quantum fluctuations ) from the outset and _ deduce _ a scalar factor of this form ( _ i.e. _ a mass ) from the coupled system .
traditionally , interactions in quantum field theory are treated as small perturbations around the non - interacting background state ( feynman diagram method , etc ) .
this is evidently applicable only to mildly non - linear theories , _
e.g. _ , theories in which the coupling constant , @xmath66 , is small , as for instance in quantum electrodynamics .
for a theory of masses , however , the exact formulation is required , as a truncated perturbative series does not capture the kind of non - linearity we are interested in .
furthermore , the very act of ( perturbative ) renormalization destroys all information about any possible underlying mechanism for mass generation , as the real ( measured ) masses must be taken as experimental input .
so , it seems obvious that a perturbative _ fundamental _ understanding of masses is impossible .
one has to take into account that nature seems to work wholesale and does `` everything at once '' , not piecemeal as implied by feynman diagrams ( _ i.e. _ by a perturbative expansion of an otherwise hitherto intractable problem ) .
it is well known that nonlinear theories may have particle - like solutions .
there are many known solitary wave solutions " , where the normal dispersive effect is exactly counterbalanced by a nonlinear focusing effect , giving waveforms that are unaltered . if the solitary waves also are unaffected by collisions with other solitary waves they are called solitons " .
soliton solutions thus in many ways act like particles .
there are many known examples of analytical soliton solutions for 1 + 1 dimensional systems ( korteweg - de vries , sine - gordon , etc ) , much fewer for 2 + 1 , and none for relativistic 3 + 1 dimensional systems with non - trivial soliton scattering .
as quantized solitons arising in relativistic field theories possess many of the attributes of particles , such as mass , charge and spin @xcite , it would be very tempting to identify elementary particles with exact solitary wave solutions to the 3 + 1 dimensional quantum field theory of the standard model , or its ( unknown ) dual @xcite , and stable elementary particles with soliton solutions to the same equations .
so far there is no integrable model with solitons respecting the symmetries in minkowski space - time .
( an example of an integrable quantum field theory , believed to bear a close similarity to the four - dimensional yang - mills theory of the standard model , is the two - dimensional nonlinear sigma model . )
numerical studies of non - integrable models show that solitons do scatter , soliton - antisoliton collisions lead to annihilation , the energy dissipating into wave - like solutions of the linearized field equations , sometime after transient periods in which the energy localizes into wavepackets called oscillons " .
this makes it tempting to identify soliton - antisoliton annihilation as _ particle - antiparticle _ annihilation , oscillons with secondary ( semi - stable ) daughter / cascade particles , and the radiated energy as the final liberated energy from annihilation . in that vein , skyrme s old pioneering unified model of nucleons and nuclei @xcite , in terms of soliton skyrmions " , with almost no free parameters , seems a compelling inspiration ( and goal ) also for truly elementary particles .
self - interacting non - linear systems have been studied for a long time in the theory of dynamical systems , chaos @xcite and spontaneously self - organizing phenomena @xcite .
they are ideal for investigating the behaviour of systems which self - interact over and over again , hence making them useful models for our ideas about mass - generation .
moreover , there is a limiting behaviour which gives the same results for large classes of models , regardless of dynamical details ( universality " ) @xcite .
this limiting behaviour is not only qualitative , but _
quantitative_. it is regarded as a fundamental organizing principle of nature on the macroscopic scale , but could in principle also apply to the quantum world as elementary particle interactions are inherently non - linear as noted above . furthermore , the production and decay of unstable ( heavy ) elementary particles is physically a clear sign of far - from - equilibrium behavior , _
e.g. _ @xmath73 .
non - equilibrium and nonlinearity generally being regarded prerequisites for self - organization @xcite .
as the vacuum is displaced further and further from equilibrium ( _ i.e. _ becomes excited " ) , more and more states become possible , see fig 2 .
we like to picture the elementary particles as semi - localized knots " in the fields permeating space , arising because of the non - linear response of the field due to its charge(s ) .
mass would then be just another example of an emergent phenomenon so characteristic of many nonlinear systems .
for example , in solid - state physics , intrinsic localized modes " @xcite , @xcite were surprisingly discovered in the late 1980s and are now known to be typical excitations in strongly nonlinear systems . as quantum field theory is also a nonlinear system , it does not take much imagination to expect analogous phenomena to occur also in particle physics .
so instead of particle masses arising from _ extrinsic _ sources ( brute - force , linear , _ ad hoc _ higgs couplings ) they could be due to _ intrinsic _ sources ( automatic , nonlinear , physical self - interactions ) .
it is perfectly possible to get phenomena of great complexity even with a remarkably simple underlying setup .
particle masses may be just yet an example of this very general phenomenon .
the self - interaction mapping " can not be linear , as this would lead to infinite ( exponentially increasing ) or vanishing ( exponentially decreasing ) masses .
one of the simplest nonlinear equations fulfilling these criteria is the celebrated `` logistic mapping '' @xmath74 which still leads to surprisingly complex dynamics @xcite and universality @xcite . in order to indicate how
masses could be constructed in such a setting , we use the logistic mapping as a toy model " .
the true dynamics , however , is believed to be much more complicated , but the _ qualitative _ features of the true mechanism could be similar , as even models governed by partial differential equations ( formally @xmath75-dimensional dynamical systems ; equivalent to dynamical systems of infinitely many coupled equations ) can , very surprisingly , be well - described by low- or even one - dimensional nonlinear dynamical systems @xcite .
dissipation bleeds a complex system of many conflicting motions , eventually bringing the behavior of many dimensions down to one " @xcite .
we take @xmath76 to be proportional to the mass generated by the self - interaction of the field with itself ( @xmath77 ) . if we again use the fermi - constant ( the only dimensionful coupling constant in the standard model ) @xmath78
is a dimensionless mass parameter .
the discreteness in time could be justified by the possible existence of a smallest time - interval ( _ e.g. _ the `` planck time '' ) but also from a pragmatic point of view as it makes calculations simpler .
the non - linearity parameter is a function of the dimensionless couplings , @xmath79 and we start with @xmath80 due to quantum fluctuation .
the physical justification of eqs .
( 14 ) , ( 15 ) comes from fig .
1 . each loop contributes @xmath81 , but at the same time nature can borrow " only a finite energy from the vacuum ( the carrying capacity " in the logistic equation ) taken as @xmath59 in this case .
so , physically , the first ( loop ) term in eq .
( 14 ) leads towards the normal diverging self - mass contribution , whereas the second term tends towards starvation" / suppression .
heisenberg s uncertainty principle , due to the finiteness of @xmath82 and the discrete time - step , means only a finite energy is available from the vacuum .
the logistic mapping , for small @xmath83 , has the behavior : @xmath84 .
this would then be interpreted that too weak self - interactions will result in strictly massless particles ( _ e.g. _ gravitons ) . for @xmath85 , _
i.e. _ point _ attractors _ , giving single asymptotically stable solutions for intermediate couplings . @xmath86 give multiple solutions , perhaps identifiable as generations . as the coupling @xmath83 increases , previously stable solutions become unstable ( repellors )
while the new bifurcation solutions become new stable states ( attractors ) .
stable particles could thus correspond to fixed points ( @xmath87 ) which are _ attractors _ while unstable particles may correspond to fixed points which are _
repellors_. @xmath88 chaotic solutions , interspersed with infinitely many narrow windows of multiple solutions .
this might be connected to the suggestion that nature is chaotic at the basic fundamental level , but goes through self - ordering at higher ( observable ) levels @xcite .
when the coupling decreases , we see that we get unification " , not necessarily of forces , but of _ masses _ , implying that there exists only _
one _ charged and massless lepton as @xmath89 . furthermore ,
if the dynamics is governed by a gut / string - like theory , with only one primordial coupling parameter , quarks and leptons are _ the same _
( apart from hitherto unexplained quantum numbers ) .
as @xmath90 for the logistic equation , we see that elementary particles , in this simple model , can not have arbitrarily large masses , but that @xmath91 gev .
this would mean that the top quark ( @xmath92 ) is presumably the heaviest elementary particle that exists .
if we assume the model in section 2 to also apply for gauge bosons , we see that only electromagnetism ( quantum electrodynamics ) has truly infinite reach ; as the photon carries no charge it has no self - interaction and its physical mass remains zero . the strong force ( quantum chromodynamics )
should disappear exponentially at sufficient distances due to the non - zero effective physical mass of its force carrier particles , the gluons , due to their self - interactions .
the range can be estimated by the non - massless theory potential @xmath93 , giving @xmath94 .
this gives for the gluon with bare mass zero ( preserving gauge - invariance in the lagrangian ) , but physical mass @xmath95 , the value @xmath96 1 fm .
this explains why qcd is only active within nuclei , although the bare gluon mass @xmath71 naively would give infinite reach within the standard model as its coupling to the higgs is zero - despite what many think , this problem has never been solved @xcite .
the very massive gauge bosons @xmath97 and @xmath98 evidently pose a problem . in our scheme @xmath99 and @xmath100 , which of course is ruled out .
there are other theoretical expectations however , which point at the possibility of @xmath101 and @xmath102 being composite particles , in which case their intricate inner structure would set the mass scale @xcite , @xcite .
if we extrapolate outside the standard model to also include gravity , the mass of the graviton , the hypothetical carrier of at least weak gravity @xcite , should be scale dependent .
the mass of a soft " graviton would still be zero for all practical purposes , while the mass of a highly energetic graviton should be proportional to its frequency due to self - interaction with its own field ( as the `` gravitational charge '' , _
i.e _ mass , increases with increasing energy ) .
gravity would then be described by a gauge theory similar to the weak interaction ( with massive spin-2 quanta instead of the spin-1 quanta of the weak theory ) .
as both the fermi coupling constant @xmath53 and newton s gravitational constant @xmath103 have dimensions @xmath104 in units where @xmath105 , it is not inconceivable that the incorporation of massive intermediate field quanta , where the mass is _ not _ put into the lagrangian by hand " , could render perturbative quantum gravity tractable , just as in the case of the weak interaction . the difference being that the mass of the graviton would be scale - dependent , whereas the @xmath101 and @xmath102 in the standard model are taken to have constant masses .
( actually , _ all _ quanta would be affected in a similar way , as they all carry `` gravitational charge '' , _
i.e. _ , energy - momentum , possibly giving an effective cut - off to loop integrals , rendering all quantum field theories ultraviolet finite . )
this would alter , in a dramatic way , the behaviour of gravity at short distances , giving an asymptotic effective coupling which tends to zero for very high energies , softening the ultraviolet divergences .
it would also give the right limiting behaviour of classical ( very many soft , low - energy gravitons ) gravity , as only the `` soft '' , essentially massless , gravitons can escape an appreciable distance to contribute to the long range gravitational field .
the effective ( non - relativistic ) gravitational potential would then be modified to something like @xmath106 where @xmath107 is the frequency dependent effective mass of the graviton .
it would possibly also modify the _ very _ long range classical behaviour of gravity , making it essentially non - infinite in reach , altering the long - range @xmath108-dependence , in accordance with some modifications proposed to explain galactic dynamics without `` dark matter '' and cosmological expansion without `` dark energy '' .
we have seen that it might be possible to understand , both qualitatively and quantitatively , much of the fundamental particle spectrum in a pragmatic way , without turning to untested `` exotic '' physics .
we have not proven anything , but then again physics is not about proofs but about correspondence with empirical facts .
still our scenario seems so simple , natural and physically compelling that we believe it to probably contain at least part of the real mechanism .
however , more refined and sophisticated models of this kind are needed before anything conclusive can be said about this proposition , although we feel it is a step in the right direction to be able to actually _ derive _ masses from fundamental principles . as essentially nothing is known today about why elementary particle masses have the values observed ,
each small step towards a resolution of this puzzle is worthwhile .
the very first small steps towards the ideas presented in this article were described in @xcite .
nieto _ et al .
_ , _ theoretical motivation for gravitation experiments on ultra - low energy antiprotons and antihydrogen _ , third biennial conference on low - energy antiproton physics , leap 94 , eds . g. kernel _ et al . _
( world scientific 1995 ) , p. 606 .
h. fritzsch , _ the mass hierarchy of leptons and quarks as a new symmetry _ , elementary particles and the universe : essays in honor of murray gell - mann , ed .
schwarz ( cambridge university press 1991 ) . | in contemporary particle physics , the masses of fundamental particles are incalculable constants , being supplied by experimental values . inspired by observation of the empirical particle mass spectrum , and their corresponding physical interaction couplings , we propose that the masses of elementary particles arise solely due to the self - interaction of the fields associated with the charges of a particle . a first application of this idea is seen to yield correct order of magnitude predictions for neutrinos , charged leptons and quarks .
we then discuss more ambitious models , where also different generations may arise from _
e.g. _ self - organizing bifurcations due to the underlying non - linear dynamics , with the coupling strength acting as non - linearity " parameter .
if the model is extended to include gauge bosons , the photon is automatically the only fundamental particle to remain massless as it has no charges .
it results that gluons have an effective range @xmath0fm , physically explaining why qcd has finite reach . |
entanglement plays an important role in respect that it is a valuable resource in quantum information processing such as quantum teleportation1 , quantum computation@xcite and quantum cryptography@xcite , _
etc_. several schemes have been proposed to prepare purified and distilled entangled state both theoretically and experimentally@xcite .
although the interaction between a quantum system and its surroundings can result in inevitable decoherence of the quantum system , people have recognized that we can employ the interaction to generate entanglement@xcite .
the two - atom entangled states are widely studied in cavity qed6,7,8,9,10 . in cavity qed ,
the dissipation in the model of atoms interacting with magnetic field generally includes two aspects : the cavity leakage through which the intra - cavity magnetic field can exchange information with its environmental noise , the atomic spontaneous emission that is induced by vacuum fluctuation effect . in the sense of using the impact of environmental noise ,
the noise - assisted entanglement schemes have been put forward by many authors@xcite .
plenio and co - work have developed schemes that involves continuous monitoring of photons leaking out of the cavity to entangle atoms one of which is initially exicted@xcite . in ref .
[ 10 ] , the author studied the interaction of a thermal field with a two - qubit system that initially prepared in separable states .
they demonstrated that entanglement of atom - atom can arise depending on initial preparation of the atoms .
also in ref .
[ 11 ] , the entanglemet of atom - atom can be generated through interaction of atoms with cavity mode coupled to a white noise .
their entanglement can be maximized for intermediate value of noise intensity and initial value of spontaneous rate . in these studies ,
the couplings of atoms to field are confined to be equal .
in fact , the coupling rate @xmath1 between atomic internal levels and the cavity mode depends on the atom s position @xmath2 @xcite .
the atoms can not be localized precisely even by employing cooling technology and trapping potential schemes .
so , it is practically necessary to address the question : how will the entanglement be when two atoms differently couple to a single model field ?
in ref.@xcite , our gruop had shown that different couplings can really assist the induce of entanglement in one - photon process .
on the other hand , the atomic spontaneous emission rate is also related to atoms s position@xcite . in real experimental scenario , the atoms s position @xmath2 not only dominates the atom s coupling strength to the field , but also determines the amount of atomic spontaneous emission rate .
it has already been reported that the resonant cavity which was made of two spherical niobioum mirrors can enhance or suppress single atomic spontaneous emission by adjusting atom position @xmath3 ( the distance from median plane of cavity)@xcite . but theoretically , atomic spontaneous emissions have been assumed to be equal or even been ignored , and the spontaneous emission has been disliked because of its impact on the entanglement@xcite . up to now
we have not found the study that two atoms spontaneous emission rates are not the same .
addition to that , the two - photon process is a kind of important one which may show different properties from the case of one photon in quantum information processing , for example , it has been found that the atom - atom entanglement induced by thermal field in two - photon process is larger than that in one - photon process@xcite . in this paper ,
considering the two - photon process , we aim to study the two atoms simultaneously interacting with a single - mode cavity field with different couplings and different spontaneous emission rates .
we find that in two - photon process we indeed can employ the different couplings to produce the the atom - atom thermal entanglement . if the atoms spontaneously emit inevitably , the different spontaneous emission rates is utilizable in generating thermal entanglement .
we also investigate the effect of the cavity leakage . to the initial atomic state @xmath4
, the cavity dissipation should be supressed as possible as we can , but to the initial atomic state @xmath0we can keep a slight leakage to relieve the restriction of interaction time so that we can obtain a large and steady entanglement .
the two two - level identical atoms ( atom @xmath5 and atom @xmath6 ) are supposed to interact with a single mode cavity field which is in a thermal equilibrium with its environment characterized in terms of an mean photon number @xmath7 , and @xmath8 is the environmental temperature .
we assume the excited atom can transit from its upper state to its lower state and emit two photons . so that , the atomic transition frequency @xmath9 doubles the field frequency @xmath10 .
the _ hamiltonian _ under the rotating wave approximation is @xmath11 where @xmath5 and @xmath12 represents annihilation and creation operator of cavity mode respectively .
the operators @xmath13 and @xmath14 denote atomic transition operators of atom @xmath15 .
the coupling constant for two - photon transition between atom @xmath15 and the cavity mode is @xmath16 . in the interaction picture ,
the _ hamiltonian _ is @xmath17 for the sake of the two couplings diversity , the following transformation is preferred @xmath18 where the @xmath19 is in the range of @xmath20 and @xmath21 . for generality , we assume the intra - cavity system can exchange information with thermal environment due to cavity dissipation and atomic spontaneous emission.the time evolution of the global system ( atoms+cavity mode ) is governed by the master equation @xmath22+l(\rho ) \text{.}\ ] ] the _ liouvillean _ that describes the atomic spontaneous emission and the interaction of the cavity mode with the thermal environment in a leaky cavity is written as@xcite @xmath23 \\ & & -\kappa ( \bar{n}+1)(a^{+}a\rho + \rho a^{+}a-2a\rho a^{+ } ) \notag \\ & & -\kappa \bar{n}(aa^{+}\rho + \rho aa^{+}-2a^{+}\rho a)\text { , } \notag\end{aligned}\ ] ] where the terms including @xmath24 in @xmath25 are interpreted as the coupling strength of cavity mode to the external thermal field , @xmath26 is the spontaneous emission rate of atomic @xmath27 . since @xmath28 can be different from @xmath29 , we adopt the transformation similar to eq .
3 @xmath30 the wootters concurrence that has been proved effective in presenting the entanglement degree of two qubits is written as@xcite @xmath31 where the @xmath32 are non - negative real square roots of the eigenvalues of the _ hermitian _ matrix @xmath33 in decreasing order with @xmath34 .
no matter what @xmath35 stands for a pure or a mixed entangled state , wootters concurrence is available . the amount of entanglement measured by concurrence on the basis of different initial atomic states will be numerical calculated in next two sections .
we assume that the single mode cavity field is initially in a thermal field state .
due to the cavity leakage when the cavity is in a thermal equilibrium with its environment , the cavity field is in a mixture of fock states .
so , the cavity field initially takes the form @xmath36 firstly , we study the effect of relative coupling difference @xmath19 on the two - atom entanglement when two - photon process is involved .
the chosen parameters are @xmath37 , @xmath38 , @xmath39 and @xmath40 .
we have cut off the intra - cavity photon number at a value of @xmath41 which is precise enough in respect that @xmath42 is a decreasing function of photon number .
fig.1a shows the entanglement as a function of relative coupling difference @xmath19 and time @xmath43 in the case of the two atoms are initially in @xmath44 , and fig.1b is the same as fig .
1a except that the two - atom are initially in @xmath4 . in ref.[14 ] , the authors could not find the entanglement induced by thermal field in two - photon process when the initial atomic state is @xmath44 if two couplings are equal .
1a also shows there is no entanglement when @xmath45 .
but if @xmath46 in some region one can find entanglement .
so , in two photon process the different couplings can also benefit to produce entanglement . comparing fig.1a with fig.2 of ref.[14 ]
, we find that the entanglement in one - photon process ( fig.2 of ref.[14 ] ) exists in some discontinuous small areas in terms of @xmath19 and @xmath43 , i.e. , for different @xmath19 entanglement may appear in different interval of time , however , fig .
1a shows that the entanglement appears in some continuous regions , that is to say , in the relative slowly varying region of time the entanglement keeps its value even the relative large change of @xmath19 .
this property will be more obvious when the two - atom are initially in @xmath4 , which is shown in fig .
the behavior that the entanglement varies with @xmath19 and @xmath43 is very interesting , and the entanglement can exist in almost the same interval of time for different @xmath19 .
for example , in the region @xmath47 , the entanglement increases to a maximum slightly with the increasing of @xmath19 from @xmath20 to @xmath48 , then it decreases to zero . in other words , if we control the interaction time in the interval @xmath47 , we need not have to care much about whether the two atoms are in the same position or not .
so , in two - photon process it is experimentally not necessary to control the position precisely , especially when the initial atomic state is @xmath4 .
then , we consider the effect of different spontaneous emission rates of two - atom on the amount of entanglement . we show the typical result of atom - atom entanglement as a function of the difference between two emissions and the noise intensity ( mean photon number of thermal environment ) in fig .
2a and fig .
2b corresponding to atomic initial states @xmath49 and @xmath50 respectively .
the chosen parameters are @xmath40 , @xmath37 , @xmath51 , @xmath52 and @xmath53 in fig .
2a and @xmath40 , @xmath37 , @xmath51 , @xmath54 and @xmath53 in fig .
2b . from fig .
2 , we see that the amount of entanglement when difference of two emissions equals to zero , i.e. @xmath55 , is not the best case of atom - atom entanglement . the maximum value of entanglement is monotonously increased by increasing the relative difference of two spontaneous emissions @xmath56 . for example , the entanglement when @xmath57 in fig .
2a is about @xmath58 times of that when @xmath59 , and in fig .
2b , the entanglement when @xmath60 is even enhanced to be about @xmath61 times of that when @xmath59 .
and fig.2a shows that the entanglement decreases monotonously with the increasing of mean photon number which is also observed in ref.[14 ] .
one can also observe that the entanglement can be increased by increasing mean photon number in some region in fig .
this is because that the two atoms initially in @xmath50 can not be entangled when they interact with vacuum field state . with the increasing of mean photon number in some extent
, the entanglement is gradually increased to a maximum .
one can observe that the amount of entanglement with spontaneous emission is quite different from that in fig .
1(without spontaneous emission ) .
when there is atomic spontaneous emission , even this emission is very weak , the amount of entanglement will be much weakened . as mentioned above , in any experimental scenario
, the atomic spontaneous emissions can hardly be all kept as zero .
therefore , any entanglement that has been realized experimentally is in fact smaller than theoretical result of ideal model . to investigate the influence of atomic spontaneous emission on the atom - atom entanglement , the authors in ref .
[ 12 ] assume two atoms have a same spontaneous emission @xmath62 in a vacuum cavity .
their results show that the amount of entanglement is a monotone decreasing function of @xmath63 . while , if there is inevitable spontaneous emission in experiment , the difference of spontaneous emission rates can also assist atom - atom entanglement .
we now turn to consider the situation when cavity keeps on leaking throughout the whole evolution .
fig.3 shows the atom - atom entanglement changing with cavity dissipation @xmath24 and time @xmath43 .
3a and fig .
3b are corresponding to the initially atomic state@xmath4 and@xmath64 respectively .
the chosen parameters in both cases are @xmath65 , @xmath66 and @xmath67
. when the two - atom are initially in @xmath4 , the amount of entanglement is a monotone decreasing function of cavity decay . with the cavity dissipation increasing ,
the entanglement decreases rapidly .
it denotes that we should depress the cavity dissipation as possible as we can if the initial atomic state is latexmath:[$% @xmath64 , a slight increasing of @xmath24 makes the period of entanglement disappears and futhermore benefits to generate relative steady and strong entanglement . although the entanglement may decrease slightly as time evolution , we still can employ the non - period to relieve the restriction of interaction time . in experiment
, precisely controlling interaction time is still very difficult . while , it will be not necessary to precisely control the interaction time by employing the slight cavity dissipation .
thus , the dissipation of the cavity is not always bad to the atom - atom thermal entanglement . in some cases such as the initial atomic state @xmath69
, the cavity dissipation is utilizable .
to sum up , when two - photon process is involved , we study two atoms simultaneously interact with the thermal field under different couplings and different spontaneous emission rates . to different initial atomic state ,
the different couplings assist to produce the the atom - atom thermal entanglement in two - photon process .
this entanglement is more regular than that of one photon process in sense that in some time intervals the entanglement can survive when difference of two couplings varies in a large range . if the atoms spontaneously emit inevitably , the different spontaneous emission rates is utilizable in generating thermal entanglement .
the different spontaneous emission rates can be realized experimentally by localize different atoms in different places in a same f - p cavity .
we also investigate the effect of the cavity leakage . to the initial atomic state latexmath:[$% d. bouwmeester , a. ekert , and a. zeilinger , quantum computation ( springer verlag , berlin , 2000 ) ; stephen d. bartlett and william j. munro , phys .
* 90 * , ( 2003 ) 117901 ; n. yoran and b. reznik , phys .
* 91 * , ( 2003 ) 037903 .
j. a. jones , m. mosca , and r. h. hansen , nature ( london ) * 393 * , ( 1998 ) 344 ; christoph simon and jian - wei pan , phys .
lett . * 89 * , ( 2002 ) 257901 ; w. dr and h .- j .
briegel , phys .
lett . * 90 * , ( 2003 ) 067901 ; antonio acin , lluis masanes , and nicolas gisin , phys .
lett . * 91 * , ( 2003 ) 167901 ; mitsuru hamada , phys .
a * 68 * , ( 2003 ) 012301 .
2a : two - atom entanglement versus difference of two spontaneous emission rates @xmath56 and mean photon number @xmath74 for atomic initial state @xmath49 , @xmath37 , @xmath51 , @xmath40 , @xmath53 and @xmath52 . | we study the system that two atoms simultaneously interact with a single - mode thermal field via different couplings and different spontaneous emission rates when two - photon process is involved .
it is found that we indeed can employ the different couplings to produce the atom - atom thermal entanglement in two - photon process .
the different atomic spontaneous emission rates are also utilizable in generating thermal entanglement .
we also investigate the effect of the cavity leakage . to the initial atomic state @xmath0a slight leakage can relieve the restriction of interaction time and we can obtain a large and steady entanglement .
pacs number : 03.67.-a , 03.67.-hz , 42.50.-p |
trapped ions are currently being studied extensively with regard to their possible use in the realisation of a prototype quantum computer @xcite .
laser - cooled trapped ions offer distinct advantages over many other proposed quantum technologies for this purpose . as a result , much work in the area of quantum information science has focused on the use of miniature radiofrequency ( paul ) traps ( for single ions ) or linear radiofrequency traps ( which allow the preparation of stationary strings of ions ) . in both cases , strong confinement to the lamb - dicke regime may be obtained ( e.g. @xcite ) .
unfortunately , the low - temperature limit of the paul trap can be affected by the micromotion of the trapped ions .
this arises as a result of the applied radiofrequency trapping field and can not be completely avoided , although the micromotion is generally minimised by the application of potentials to compensation electrodes .
decoherence rates may be limited by the presence of the micromotion or by the small size of the trap electrodes , which allow image charge and patch potential effects to become significant @xcite .
ions in a penning trap have no micromotion because the applied electric and magnetic fields are both static .
furthermore , since the radial confinement is given by the magnetic field , strong confinement does not require small electrodes as it does in the case of radiofrequency traps .
however , the nature of the motion in the radial plane of the penning trap complicates the laser cooling process ( e.g. @xcite ) . in the penning trap ,
the radial motion ( perpendicular to the magnetic field @xmath1 ) of an ion ( mass @xmath2 , charge @xmath3 ) is described by a combination of two circular orbits : a fast cyclotron motion at the modified cyclotron frequency @xmath4 and a slow magnetron motion at @xmath5 .
these frequencies are given by @xmath6 @xmath7 where @xmath8 @xmath9 is the true cyclotron frequency and @xmath10 is the axial oscillation frequency ( all formulae for frequencies have been quoted in angular frequency units ) . here
@xmath11 is the applied voltage and @xmath12 is a parameter determined by the diameter of the ring electrode ( @xmath13 ) and the separation of the endcaps ( 2@xmath14 ) .
it can be shown that it is difficult to decrease simultaneously the amplitudes of both the magnetron and cyclotron motions with conventional doppler laser cooling @xcite .
this is because the magnetron motion is unstable : its total energy is negative .
therefore energy must be supplied in order to reduce the magnetron radius , which can be achieved only when the laser is placed at the point where the magnetron rotation moves the ions in the same direction as the laser photons .
even so , the cooling of the cyclotron motion is considerably more efficient than that of the magnetron motion . as a result
, the final amplitude of the magnetron motion for a laser - cooled ion in a penning trap is not as small as might be expected from the standard doppler temperature limit , limiting the degree of localisation of the ion that can be achieved by laser cooling .
this is in contrast to the radiofrequency trap , where very tight localisation may be achieved . in some cases ,
even cooling to the lamb - dicke regime is possible with doppler cooling , without the need for more advanced laser cooling techniques @xcite . for large numbers of ions in a penning trap , the ions form into a relatively dense cloud with a size limited by the effects of space charge .
this cloud is effectively a one - component non - neutral plasma and the dynamics are best treated in terms of plasma parameters . under these conditions ,
the plasma has a well - defined size and shape and it rotates as a whole at a rate which depends on its density ( this rotation is analogous to the magnetron motion of a single particle ) .
laser cooling can be used to cool and compress the cloud , but this process is limited by the dynamics of the trap @xcite . however
, the situation can be improved by imparting an external torque to the plasma .
one way of doing this is to use the rotating wall " technique .
this increases the plasma density by locking the rotation of the plasma to an external drive field and increasing this rotation frequency to half the true cyclotron frequency , which corresponds to maximum plasma density @xcite . for the case of a single ion in a penning trap
, the motion is described using the single - particle parameters introduced above .
this treatment also holds well for a few ions , when the cloud is not large enough to behave as a plasma . in our experiments we always studied a small number of ions , so in the rest of this paper we use the single - particle treatment of the ion motion .
axialisation is the process by which particles in a penning trap are driven towards the central axis of the trap @xcite .
it is also sometimes referred to as azimuthal quadrupole excitation or as sideband excitation of the cyclotron motion .
it occurs when the efficient cooling of the cyclotron motion is effectively extended to the magnetron motion by means of a coupling between the two .
this results in a reduction in the amplitudes of both motions .
axialisation has been used in conjunction with buffer gas cooling to increase storage times in penning traps and to increase the density of ion clouds @xcite .
a related technique is also used in precision mass measurements @xcite .
combined with laser cooling , one expects very low effective temperatures to be reached using axialisation .
the process of axialisation is driven by a radial quadrupole field at @xmath15 generated by a ring electrode split into four quadrants .
this couples the magnetron and cyclotron motions .
the quadrupole geometry is needed in order to drive the correct combination of motional frequencies ; other geometries select different combinations @xcite . in the frame which rotates at @xmath16 , in which the magnetic field is effectively eliminated by the rotation , the applied field generates an additional _ static _ quadrupole potential which couples the clockwise and anticlockwise motions in this frame .
these are both at angular frequency @xmath17 and correspond to the cyclotron and magnetron motions in the laboratory frame . in the absence of additional cooling
, the coupling gives a slow cycling of the motion between the cyclotron and magnetron motions .
this effect has been employed in precision mass measurements @xcite . in a quantum mechanical picture ,
each excitation at the sum frequency can be thought of as causing the quantum number of the cyclotron motion to increase by 1 and that of the magnetron motion to _ decrease _ by 1 .
this reduces the magnetron radius @xmath18 and increases the cyclotron radius @xmath19 .
the excess energy in the cyclotron motion is then rapidly removed by the laser cooling .
the reverse process also occurs but this is generally weaker ( see below ) .
an intuitive picture of the axialisation process is illustrated schematically in figure [ figure1 ] , where the axialisation drive produces the transition @xmath20 and then the laser cooling , via excitation to the excited state , effectively produces the transition @xmath21 .
the overall change , for the particular sequence of events illustrated , is therefore to reduce the magnetron motion quantum number by 1 .
note , however , that because the sidebands on the optical transition due to the ion motion are not resolved in our case , the actual change of @xmath22 in a particular optical excitation may be different from 1 .
motional energy levels of an ion in the penning trap ( not to scale ) .
note that the cyclotron energy is positive but that of the magnetron motion is negative .
the solid bold upward arrow shows transitions driven by the axialisation drive at @xmath15 , and the downward dashed arrow represents the effect of a cycle of laser cooling reducing the quantum number of the cyclotron motion . in our experiments , @xmath4 and @xmath5
are typically @xmath23 600 khz and @xmath23 30 khz respectively.,width=302 ] the process can be described classically by the following equations : @xmath24 @xmath25 where @xmath26 and @xmath27 are the cooling rates for the two motions and @xmath28 is the coupling rate due to the rf axialisation field . assuming that the diameter of the ion s motion is initially smaller than the laser beam , various solutions to the equations can be found : in figure [ fig2 ] we show simulations of ( a ) oscillation between magnetron and cyclotron motions when the two damping rates are zero ; ( b ) axialisation when @xmath29 ; and ( c ) a stable orbit with finite radius when @xmath30 .
this requires the cyclotron and magnetron damping rates to be of opposite sign : this situation arises in the penning trap when the laser beam passes through the centre of the trap .
simulation of ion motion in the presence of the axialisation drive : ( a ) with no cooling the ion cycles between cyclotron and magnetron motions ; ( b ) with weak laser cooling the ion spirals towards the centre of the trap ; ( c ) with strong laser cooling a stable orbit may be obtained.,title="fig:",width=302 ] simulation of ion motion in the presence of the axialisation drive : ( a ) with no cooling the ion cycles between cyclotron and magnetron motions ; ( b ) with weak laser cooling the ion spirals towards the centre of the trap ; ( c ) with strong laser cooling a stable orbit may be obtained.,title="fig:",width=302 ] simulation of ion motion in the presence of the axialisation drive : ( a ) with no cooling the ion cycles between cyclotron and magnetron motions ; ( b ) with weak laser cooling the ion spirals towards the centre of the trap ; ( c ) with strong laser cooling a stable orbit may be obtained.,title="fig:",width=302 ] if the coupling rate @xmath28 is small and the damping rates are of opposite sign , then @xmath31 and the orbit starts to expand . because of the finite size of the laser beam , the orbit then takes the ion out of the laser beam for some of the time .
the interaction with the laser therefore weakens , reducing the values of the damping rates .
this process continues until the above equality in ( c ) is satisfied , resulting in a stable orbit whose size depends on the laser parameters . note that with laser cooling the values of the damping coefficients can be varied by changing the laser frequency and position @xcite .
the axialisation process was previously demonstrated with buffer - gas cooling where the relative damping rates are fixed and do not vary with the amplitude of the motion .
the appearance of the stable orbit is therefore a feature unique to laser cooling .
our set - up consists of a penning trap with an internal diameter of 10 mm located between the pole pieces of an electromagnet generating a field of roughly 1 t @xcite .
the ring electrode is split into four segments for application of the quadrupole axialisation drive .
the trapped magnesium ions are doppler cooled with laser radiation tuned close to the resonance line at 280 nm .
the natural linewidth of this transition is 43 mhz ( giving a theoretical doppler limit of 1 mk ) so the motional sidebands are not resolved ( all the ion oscillation frequencies are less than 1 mhz ) .
the fluorescence is detected by a photomultiplier which feeds pulses to a multichannel scaler ( mcs ) ( to measure the fluorescence rate ) or to a time to amplitude converter ( tac ) and multichannel analyser ( mca ) ( for photon correlation studies ) .
some of the light can be imaged with a magnification of @xmath32 onto an iccd having a pixel size of 13 @xmath0 m .
if the fluorescence signal is monitored by the photo - multiplier , an increase is observed immediately when the axialisation drive is applied , as the size of the cloud is reduced and the ions are cooled more strongly @xcite .
the iccd images show an immediate reduction in the size of an ion cloud when the axialisation drive is turned on , showing that compression of the cloud has been achieved .
it is then possible to optimise the conditions to improve the axialisation , which further reduces the size of the cloud .
figure [ fig3 ] shows an image of an optimised single ion .
the measured size of the image is determined to be of the order of 20 @xmath0 m , which is comparable to the finite resolution of the camera and imaging system .
the equivalent temperature for the single ion is of order 10 mk @xcite .
this should be seen as an upper limit because of the finite resolution of the camera and imaging system .
the observed shape of the cloud depends critically on the axialisation parameters , as can be seen in figure [ fig4 ] , which shows the size and shape of an ion cloud for various detunings of the rf drive , close to the cyclotron frequency .
this demonstrates that axialisation is most effective at the true cyclotron frequency .
image of a single ion with optimal axialisation and laser cooling.,width=302 ] images of a small axialised cloud with different axialisation drive frequencies close to resonance .
the frequencies range from 621 to 631 khz from left to right in 2 khz steps ; the fourth trace is at the true cyclotron frequency.,width=302 ]
we have measured the cooling rates for the motions of ions in a small cloud using an rf - photon correlation technique ( see @xcite ) . by driving the ions with an additional , weak rf dipole field close to a motional resonance frequency , the phase difference between this dipole drive and the resulting motion can be found .
moving through the resonance , this phase difference changes by @xmath33 ( as for any driven harmonic oscillator ) .
the frequency width of this curve gives the cooling rate ( @xmath34 ) directly .
two examples ( for the magnetron motion ) are shown in figure 5 . in earlier experiments
@xcite we measured this rate for both cyclotron and magnetron motions and found the magnetron cooling rate ( @xmath27 ) to be very low , as expected from calculations @xcite .
measured damping rates for the magnetron motion were less than 100 hz , whereas the cyclotron damping rate ( @xmath26 ) for magnesium was generally a few khz , depending on the laser parameters and the number of ions .
we have now measured the rate in the presence of axialisation ( see figure [ fig5 ] ) and we find that the magnetron cooling is enhanced due to the axialisation , confirming that the strong cooling of the cyclotron motion is extended to the magnetron motion by this technique . without the axialisation
we find a cooling rate lower than 60 hz ( similar to our earlier measurements ) , but this increases to over 300 hz after optimisation of the axialisation .
the increase in cooling rate varies approximately linearly with the drive amplitude .
these measurements were not performed on a single ion , and the rates are expected to be significantly higher for single particles .
plots of the phase of the driven magnetron motion for a small cloud of ions as a function of frequency of the drive close to the magnetron frequency .
( a ) with no axialisation present ; ( b ) with axialisation present at an amplitude of 1.5 v.,width=302 ]
when a relatively large cloud of ions was loaded into the trap , one interesting phenomenon that was observed was the production of large stable orbits , when the damping is strong ( case ( c ) in section 2 ) .
this is illustrated in figure [ fig6]a .
a large circular orbit results in two areas of fluorescence , where the orbit interacts with the laser beam .
the observation direction in our experiment is perpendicular to both the trap axis and the laser beam .
the iccd images alone can not distinguish between a single cloud moving coherently with a large magnetron radius ( i.e. , a large amplitude motion of the centre of mass of the cloud ) and a rotating ring of ions . in an attempt to resolve this issue ,
photon - photon correlation was used . in this technique ,
consecutive detected photons are used to start and stop the tac and the resulting waiting times " are collected in the mca . as the motion takes the ions out of the laser beam altogether , the magnetron motion will appear as a strong modulation of the resonance fluorescence ( figure [ fig6]b ) .
the motion at the modified cyclotron frequency also modulates the fluorescence , as a result of the doppler effect taking the ion in and out of resonance with the laser .
taking a fast fourier transform ( fft ) of the spectrum of time intervals between photons allows information about both ion oscillation frequencies to be extracted @xcite .
observation of the stable orbit of an ion cloud subjected to an axialisation drive and strong laser cooling : ( a ) image of the cloud showing two areas of fluorescence where the orbit takes the cloud through the laser beam ; ( b ) photon - photon correlation plot ( note that the sloping background comes mainly from background scattered light ) ; ( c ) fft of trace ( b ) showing modulation mainly at @xmath35 and @xmath36 .
( the observed weaker combination frequencies arise as an artefact of this technique and do not indicate that the ions are moving at these frequencies.),title="fig:",width=302 ] observation of the stable orbit of an ion cloud subjected to an axialisation drive and strong laser cooling : ( a ) image of the cloud showing two areas of fluorescence where the orbit takes the cloud through the laser beam ; ( b ) photon - photon correlation plot ( note that the sloping background comes mainly from background scattered light ) ; ( c ) fft of trace ( b ) showing modulation mainly at @xmath35 and @xmath36 .
( the observed weaker combination frequencies arise as an artefact of this technique and do not indicate that the ions are moving at these frequencies.),title="fig:",width=340 ] observation of the stable orbit of an ion cloud subjected to an axialisation drive and strong laser cooling : ( a ) image of the cloud showing two areas of fluorescence where the orbit takes the cloud through the laser beam ; ( b ) photon - photon correlation plot ( note that the sloping background comes mainly from background scattered light ) ; ( c ) fft of trace ( b ) showing modulation mainly at @xmath35 and @xmath36 .
( the observed weaker combination frequencies arise as an artefact of this technique and do not indicate that the ions are moving at these frequencies.),title="fig:",width=377 ] normally there is no correlation between photons emitted by different ions , because of the random phase difference between the motions of the ions .
therefore the correlation trace consists mainly of a smooth background in the form of an exponential decay with a time constant determined by the overall count rate .
any correlation peaks have a maximum fractional area of @xmath37 , where @xmath38 is the number of ions .
however , if the ions are all moving coherently ( i.e. , with fixed relative phases ) , a large modulation will be seen as all photons are now correlated .
if the ions move together in a single cloud , the modulation will be mainly at twice the magnetron frequency , as the motion takes the ions through the laser beam twice each orbit .
however , @xmath38 ions moving in a ring will give modulation at @xmath38 or @xmath39 times the magnetron frequency .
the fft of our data ( figure [ fig6]c ) shows clear modulation at twice the magnetron frequency and at the modified cyclotron frequency , showing that the cloud moves as a whole . if the ion cloud were not moving as a whole , then photons from different ions would be uncorrelated , so the modulation depth would be small .
if the ions were in a ring configuration , then there would be several correlation peaks per magnetron period , which would give rise to high harmonics of the magnetron frequency in the fft plot .
note that the background of the correlation plot shown in figure [ fig6]b is mainly due to scattered light , and not to fluorescence photons from the ions .
this is because the measured fluorescence rate is very low when the orbit size is large .
we have demonstrated that axialisation can be used in conjunction with laser cooling to give very tight confinement of a single ion to the axis of a penning trap .
the axialisation technique is complementary to the rotating wall technique , which is very effective for large ion plasmas but is not applicable for small clouds of ions and especially for a single ion .
we have also shown that under certain conditions a cloud of ions in the trap will move coherently in an orbit having a large magnetron radius .
this occurs when laser cooling is combined with a strong axialisation drive .
the laser cooling rate for the magnetron motion was measured for a small cloud of ions and was found to be enhanced by the axialisation process .
this should allow a single ion in a penning trap to be cooled closer to the lamb - dicke regime than is otherwise possible in this type of trap .
the small size of the image of a single axialised ion indicates that the final temperature is much lower than that obtained without axialisation . as a result
, single ions in a penning trap may be suitable for applications in quantum information processing ; it should also be possible to prepare small strings of a few ions along the magnetic field axis of a penning trap .
although in some respects the presence of a magnetic field is a disadvantage , there are proposals for experiments for quantum information processing where the presence of a magnetic field is required @xcite .
furthermore , decoherence rates for trapped ions may be lower in the penning trap than they are in the rf trap due to the larger size of the electrodes and the absence of the rf confining potential .
in the situation where large numbers of ions are used for quantum computing , the penning trap also has the advantage that a two - dimensional array of ions can be formed and imaged @xcite .
we gratefully acknowledge the support of the epsrc and the european union ( qubits project of the qipc programme ) for this work .
99 e.g. a steane , _ rep . prog . phys . _
* 61 * , 117 ( 1998 ) c roos et al .
lett . _ * 83 * , 4713 ( 1999 ) c sackett et al . , _ nature _ * 404 * , 256 ( 2000 ) r g devoe and c kurtseifer , _ phys . rev .
a _ * 65 * 063407 ( 2002 ) r c thompson and j papadimitriou , _ j. phys .
b _ * 33 * , 3393 ( 2000 ) w m itano and d j wineland , _ phys . rev .
a _ * 25 * , 35 ( 1982 ) f diedrich et al .
a _ * 62 * , 403 ( 1989 ) w m itano et al .
a _ * 38 * , 5698 ( 1988 ) x - p huang et al .
lett . _ * 80 * , 73 ( 1998 ) g savard et al .
a _ * 158 * , 247 ( 1991 ) l schweikhard et al . , _ j. am .
mass spectrom .
_ * 4 * , 433 ( 1993 ) m k " onig et al . ,
_ int . j. mass spectrometry ion proc . _ * 142 * , 95 ( 1995 ) e a cornell et al .
a _ * 41 * , 312 ( 1990 ) l schweikhard and a marshall , _ j. am .
mass spectrom .
_ , * 4 * , 433 ( 1993 ) see , for example , g bollen , r b moore , g savard and s stolzenberg , _
* 68 * , 4355 ( 1990 ) m a van eijkelenborg et al . ,
_ int . j. mass spectrom .
ion proc . _
* 188 * , 155 ( 1999 ) h powell et al .
lett . _ * 89 * , 093003 ( 2002 ) m a van eijkelenborg et al .
comm . _ * 159 * , 169 ( 1999 ) k dholakia et al . , _ phys . rev . a _ * 47 * , 441 ( 1993 ) f mintert and c wunderlich , _ phys . rev . lett . _ * 87 * , 257904 ( 2001 ) t b mitchell et al . , _ phys . rev .
lett . _ * 87 * , 183001 ( 2001 ) | we report a study of the axialisation and laser cooling of single ions and small clouds of ions in a penning trap .
a weak radiofrequency signal applied to a segmented ring electrode couples the magnetron motion to the cyclotron motion , which results in improved laser cooling of the magnetron motion .
this allows us to approach the trapping conditions of a paul trap , but without any micromotion . using an iccd camera
we show that the motion of a single ion can be confined to dimensions of the order of 20 @xmath0 m or less .
we have measured increased magnetron cooling rates for clouds of a few ions , using an rf - photon correlation technique . for certain laser cooling conditions ,
the magnetron motion of the center of mass of the cloud grows and stabilises at a large value .
this results in the ions orbiting the center of the trap together in a small cloud , as confirmed by photon - photon correlation measurements . |
in quantum information science it is a crucial problem to develop techniques for generating entanglement among stationary qubits .
entanglement as unique feature of quantum mechanics can be used not only to test fundamental quantum - mechanical principles @xcite , but to play a central role in applications @xcite .
especially , multipartite entanglement has been recognized as a powerful resource in quantum information processing and communication .
there are two typical multipartite entangled states , greenberger - horne - zeilinger ( ghz ) and w states , which are usually referred to as maximal entanglement .
numerous protocols for the preparation of such states have been proposed @xcite .
most of them are scattering - based schemes which utilize two processes : the natural dynamic process of an always on system and the final project process carried out by a subsequent measurement .
another feature of such kind of schemes is that there are two kinds of qubits involved in : target qubits and flying qubit .
the target qubits are the main entities that will be entangled by the above two processes , which are usually stationary and can be realized by atoms , impurities , or other quantum devices .
the flying qubit is amediator to establish the entanglement among the target qubits via the interaction between them , which is usually realized by photon or mobile electron . in this sense ,
the type of interaction between stationary qubits and the flying qubit as a mediator and the transfer of the flying qubit are crucial for the efficiency of the entanglement creation . in general ,
such two processes are mutually exclusive
. the scattering between stationary and flying qubits can convert information between them , while it also reduces the fidelity of the flying qubit , which will affect the efficiency of the entanglement , especially for multi - particle system .
it is still a challenge to create entanglement among massive , or stationary qubits . in this paper
, we consider whether it is possible to use an arrangement of qubits , a spin network , to generate multipartite entanglement among stationary qubits via scattering process .
we introduce a scheme that allow the generation of the ghz and w states of stationary qubits in spin networks . in the proposed scheme ,
the flying qubit is a gaussian type single - flip moving wave packet on the ferromagnetic background , which can propagate freely in @xmath4 chain .
the stationary qubit is consisted of two spins coupled by ising type interaction with strength @xmath5 .
a single spin flip can be confined inside such two spins by local magnetic field @xmath6 to form a double - dot ( dd ) qubit .
the system of an @xmath4 spin chain with a dd qubit embedded in exhibits a novel feature under the resonance scattering condition @xmath7 , that a single - flip moving wave packet can completely pass over a dd qubit and switch it from state @xmath8 to @xmath9 simultaneously .
we show that the scattering between a flying qubit and a dd qubit can induce the entanglement between them and the operation on the dd qubit can be performed by the measurement of the output flying qubit .
it allows simple schemes for generation of multipartite entanglement , such as ghz and w states by simply - designed spin networks .
we also investigate the influence of near - resonance effects on the success probabilities of the schemes .
it is found that the success probabilities are @xmath10 @xmath11 and @xmath2 for the generation of ghz and w states , respectively .
here @xmath3 is the transmission probability amplitude for a single dd qubit and @xmath0 is the number of the dd qubits .
this paper is organized as follows . in sec .
ii the dd qubit and spin network are presented . in sec .
iii we investigate the resonance - scattering process between the flying and stationary qubits .
iv and v are devoted to the application of the resonance scattering on schemes of creating ghz and w states .
section vi is the summary and discussion .
the spin network we consider in this paper is consisted of spins connected via the @xmath12 interaction .
the hamiltonian is @xmath13 \\ & & + \sum_{i}h_{i}\sigma _ { i}^{z } , \notag\end{aligned}\]]where @xmath14 , and @xmath15 @xmath16 are the pauli spin matrices for the spin at site @xmath17 .
the total @xmath18-component of spin , or the number of spin flips on the ferromagnetic background , is conserved as it commutes with the hamiltonian . for @xmath19
, it reduces to @xmath4 spin network , which has received a wide study for the purpose of quantum state transfer and creating entanglement between distant qubits by using the natural dynamics @xcite . for @xmath20 ,
the hamiltonian describes isotropic heisenberg model . in the antiferromagnetic regime ( @xmath21 ) , a ladder geometry spin network , a gapped system @xcite , has been employed as a data bus for the swapping operation and generation of entanglement between two distant stationary qubits .
it has been shown that a moving wave packet can act as a flying qubit @xcite like photon in a fiber .
on the other hand , the analogues of optical device , beam splitter can be fabricated in quantum networks of bosonic plenio , spin and ferimonic systems @xcite . in this paper , we consider a new type of qubit , double - dot qubit , which can be embedded in such spin networks .
a dd qubit consists of two ordinary spins at sites @xmath22 and @xmath23 connected via ising type interaction in the form@xmath24when such kinds of two spins are embedded in the spin networks with @xmath25 and @xmath26 , a spin flip is confined within it and forms a dd qubit with the notations @xmath27 and @xmath28 .
we will show that such a new type of qubit has a novel feature when it interacts with another spin flip in the spin networks .
the main building block in the spin network of our scheme is the dd qubit .
it acts as a massive or stationary qubit , like atoms or ions in cavity - qed - based schemes . to demonstrate the property of a dd qubit in a spin chain
, we investigate a small system of @xmath29-site , a dd qubit connecting to two spins . in order to provide a clear exposition , we firstly assume a specific coupling configuration with @xmath7 , which leads to the following @xmath29-site hamiltonian@xmath30there is a quasi - invariant subspace with the diagonal energy being @xmath6 and under the condition @xmath31 , which is spanned by basis@xmath32the matrix of the hamiltonian in this subspace reads@xmath33\]]with eigenstates @xmath34 and eigen energies @xmath35 @xmath36 being @xmath37obviously , in the invariant subspace , such a @xmath29-site system acts as a normal @xmath38-site system . note that the dd qubit as the center of the @xmath38-site system can be in two different states @xmath8 or @xmath39 while another spin flip is at left or right site .
thus a time evolution process can accomplish the transformation @xmath40 or @xmath41 with 100% success probability .
such a feature is desirable for the quantum information processing .
it is because that , on one hand , a spin flip passing over a dd qubit can operate the qubit state ; and on the other hand , the state of a dd qubit can indicate whether there is a spin flip passing over it . a similar transformation has been proposed through the cavity input - output process in adiabatic limit @xcite .
a particular merit of the present scheme is that it is based on a natural dynamic process rather than an adiabatic process .
now we consider the dynamic process of the interaction between a moving wave packet and a dd qubit .
we embed a dd qubit into a chain as illustrated in fig .
[ fig1](a ) .
it has been shown that a single - spin - flip wave packet in the form @xcite@xmath42can propagate along a uniform spin chain without spreading approximately , where the vacuum state is fully ferromagnetic state @xmath43 . here
@xmath44 is the normalization factor , @xmath45 is the center of the wave packet at @xmath46 and @xmath47 is the number of sites of the chain . at time
@xmath3 , it will evolves to @xmath48 .
let us firstly assume that initially the qubit is in the state @xmath49 , while a wave packet of type ( [ single p ] ) @xmath50 @xmath51 is coming from the left .
similarly , we define @xmath52 @xmath53 , @xmath54 @xmath55 to denote a transmitted or reflected wave packet after scattering . in the strong local field regime , @xmath56 ,
the spin flip is firmly confined in the dd qubit . from the analysis of the above @xmath29-site system , which is called the resonant case with @xmath7 , the wave packet will pass freely through the dd qubit . comparing to the case without the embedded dd qubit , the output wave packet gets a forward shift with a lattice space , while switches the dd qubit from @xmath57 to @xmath58 , i.e. , @xmath59 in contrast , if the qubit is in state @xmath9 ,
the scattering process is @xmath60 i.e. , the incoming wave packet is totally reflected and maintains the qubit to be in state @xmath58 .
interestingly , the states of wave packet and the dd qubit are both altered through this process , i.e. , being shifted with a lattice space .
however , such a shift brings about totally different effects on the dd qubit and the wave packet , respectively : it switches the dd qubit from @xmath61 to @xmath62 , but does not alter the wave packet in the same manner as _ classical _ perfect elastic collision .
it is worthy to point out that the dd qubit and the wave packet are not entangled in such resonant case . in the case of non - resonance @xmath63 , or
initially the dd qubit and/or the incident wave packet are in a superposition states as @xmath64 where @xmath65 and @xmath66 are arbitrary coefficients satisfying @xmath67 . in practice
, the difference between @xmath6 and @xmath5 will leads to the reflection of the incident wave packet .
then the scattering process can be expressed as @xmath68 with @xmath69 .
the transmission coefficient @xmath70 is a crucial factor in the following schemes for quantum information processing . on the other hand
, the strength of the local field @xmath6 also results the spreading of the spin flip from state @xmath58 and reduces @xmath71 .
we perform numerical simulation for the scattering process in order to investigate such phenomenon .
numerical result for @xmath72 with @xmath73 is plotted in fig .
it shows that the transmission coefficient is close to @xmath74 if @xmath75 , which is feasible in practice .
it ensures that a spin network with an embedded dd qubit in a spin chain can perform the transformation ( [ tran1 ] , [ tran2 ] ) via a natural dynamic process rather than an adiabatic process .
now we focus on the practical application of resonant scattering effect on the quantum information processing .
as mentioned above , although the totally transmitted wave packet switches the state of the dd qubit from @xmath76 to @xmath58 , there is no entanglement between the dd and the wave packet arising from such a process .
however , if the incoming wave packet is not polarized in @xmath18 direction , but in the form @xmath77 where @xmath65 and @xmath66 are restricted to be real for simplicity in the following context , the entanglement between the dd qubit and the scattered wave packet can be established .
actually , the corresponding resonant scattering process can be expressed as @xmath78 the reduced density matrix of the final state in the basis @xmath79 , @xmath80 , @xmath81 , @xmath82 is@xmath83 , \]]which has concurrence @xmath84 . for @xmath85 ,
the concurrence between them reaches to the maximum @xmath86 .
a flying qubit ( [ wp_in ] ) can be generated from wave packet ( [ single p ] ) via a @xmath87-beam or @xmath88-beam splitter @xcite .
we consider the spin network with the geometry of two connected @xmath87-beam splitters .
these two @xmath88-beam splitters are characterized by @xmath89 and @xmath90 respectively , which are schematically shown in fig .
fig3(a ) . there is a dd qubit embedded in one of the two arms .
now we consider the dynamic process with the initial state being @xmath91 , where @xmath92,denotes an incoming wave packet along the left chain . in the first step , through the beam splitter @xmath93
, @xmath94 is divided into two wave packets @xmath95 and @xmath96 along @xmath97 and @xmath98 arms , respectively . in the second step ,
sub - wave packet @xmath99 passes over dd qubit and becomes @xmath100 , while sub - wave packet @xmath101 propagates along @xmath98 arm and meets @xmath102 at the joint of beam splitter @xmath90 . in the third step , wave packets @xmath103 and
@xmath101are reflected and divided by beam splitter @xmath104 , and contribute to the output wave packet @xmath105 .
then the whole process can be expressed as @xmath106 from step 2 to step 3 we have used formulas ( [ alfa in ] , [ beta in ] ) derived in appendix a. when @xmath107 is measured in the output lead , the operation @xmath108is implemented .
the success probability of this operation is @xmath109 . in the optimal case with @xmath110 @xmath111
, we can perform the operation @xmath112 by the process of resonant scattering and subsequent measurement with the success probability up to @xmath113 .
the measurement of @xmath107 can be implemented by embedding another dd qubit to record the passing of the output wave packet .
now we consider the case of multiple dd qubits embedded in @xmath97 arm , which is schematically illustrated in fig .
[ fig3](b ) .
all the @xmath0 stationary dd qubits are prepared initially in state @xmath114 .
similarly , we have @xmath115with the notation @xmath116 . in the optimal case with @xmath110 @xmath111 , we can perform the operation @xmath117by the subsequent measurement with the success probability up to @xmath113 . then by using natural dynamics and subsequent measurement ,
multipartite entangled ghz state can be generated .
this provides a simple way of entangling @xmath0 stationary qubits through scattering with a flying qubit . in the near - resonance scattering case , the transmission probability amplitude @xmath3 will effect the success probability to be @xmath118under the optimal conditions @xmath119 , @xmath120 .
note that the success probability is reduced exponentially as the number of qubits increases .
now we turn to the scheme of the generation of another type of multipartite entangled state , w state . the configuration of the spin network we utilized consists of two @xmath121-beam splitters with one dd qubit embedded in each parallel arm in the same way , as illustrated in fig . [
fig3](c ) .
we start our analysis by considering @xmath122 case with two @xmath87-beam splitters being characterized by @xmath93 and @xmath123 respectively . denoting the qubit states of two dd qubits embedded in arms @xmath97 and @xmath98 as @xmath124 and @xmath125 respectively
, the dynamic process can be written as @xmath126 in the optimal case with @xmath127 , we can perform the operation @xmath128by the subsequent measurement of the output spin flip with the success probability up to @xmath113 .
now we extend the above conclusion to @xmath0-dd qubits case . for a @xmath121-beam splitter
, we only consider the simplest case with identical hopping constant being @xmath129 between the input lead and each arm , which is schematically illustrated in fig .
[ fig3](c ) .
an incident wave packet will experience the following process . in the first step , @xmath130
is divided into @xmath0 wave packets through the beam splitter , with @xmath131 @xmath132 being the wave packet along the @xmath133th arm . in the second step ,
every sub - wave packet @xmath134 passes over the corresponding dd qubit embedded in the @xmath133th arm , and switches its state from @xmath135 to @xmath136 simultaneously . in the third step ,
all the sub - wave packets @xmath137 are reflected and divided at the node of the right beam splitter , and contribute to the output wave packet @xmath138 .
the whole process can be expressed as @xmath139from step 2 to step 3 we have used the formula ( [ l_in ] ) derived in appendix b. when @xmath107 is measured in the output lead , the operation @xmath140is implemented with the success probability @xmath141 .
then by using natural dynamics and subsequent measurement , multipartite entangled w state can be generated . in the near - resonance scattering case ,
the transmission probability amplitude @xmath3 reduces the success probability to @xmath142 as the comparison of the success probabilities of creating ghz and w states of @xmath0 qubits with the transmission coefficient @xmath143 , we plot eqs .
( p_ghz ) and ( [ p_w ] ) in fig .
[ ghz_w ] for the cases with @xmath144 , @xmath145 , @xmath146 , and @xmath147 ; @xmath122 , @xmath38 , @xmath148 , @xmath149 .
it shows that when @xmath143 is close to @xmath74 , the difference between @xmath150 and @xmath151 becomes large as @xmath152 increases .
as @xmath143 decreases from @xmath74 , the difference between @xmath150 and @xmath151 becomes smaller for fixed @xmath0 .
we have shown how a spin network can be used to generate multipartite entanglement among stationary qubits . the key of this scheme is the alternative of massive or stationary qubit in a spin network , dd qubit . the resonance scattering between a dd qubit and a sfwp , which acts as the alternative of a flying qubit in a spin network , allows a perfect transformation : an incident wave packet can totally pass through a dd qubit and switch it from state @xmath8 to state @xmath153 .
the resonance scattering condition is investigated analytically and numerically .
it shows that the resonance scattering is feasible in practice .
this ensures that through the natural dynamical evolution of an incident single - spin - flip wave packet in a spin network and the subsequent measurement of the output single - spin - flip wave packet , multipartite entangled states among @xmath0 stationary qubits , ghz and w states can be generated .
there are two merits in our scheme .
firstly , the massive or stationary qubit , dd qubit , is constructed by the element , two neighbor spins of the spin network , which is applicable to all types of the scalable multi - qubit systems .
secondly , it is based on a natural dynamic process rather than an adiabatic process .
there is certainly significant potential for spin networks to find applications in solid state quantum processing and communication .
we acknowledge the support of the cnsf ( grant no . 10874091 , 2006cb921205 ) .
in this appendix , we present the exact results for the dynamics of wave packets in beam splitters . in the work of ref .
@xcite the dynamics of a wave packet in the spin networks based on the @xmath4 model was studied .
a @xmath87-beam splitter is consisted of three uniform spin chains with coupling constant @xmath154 .
the connections between three uniform chains are @xmath155 and @xmath156 as in fig .
it has been shown that under the condition @xmath157 , an input moving wave packet @xmath158 will be divided into two wave packets @xmath159 and @xmath160 without any reflection , which can be expressed as@xmath161similarly , the inverse process also holds , i.e.,@xmath162where states @xmath158 , @xmath163 , and @xmath164 ( @xmath107 , @xmath165 , and @xmath166 ) represent the wave packets coming in ( out ) of the node along the three chains respectively .
contrarily , for two wave packets along the two arms , which interference destructively at the node , we have @xmath167combining the above eqs .
( [ a(1 ) ] and [ a(2 ) ] ) , we have@xmath168and@xmath169then for a given incident wave packet along any branch of a @xmath87 beam splitter , the probability amplitudes of all the output wave packets can be obtained exactly .
now we consider a @xmath121 beam splitter consists of one chain of length @xmath170 and @xmath0 arms of length @xmath171 with identical connecting coupling strength @xmath129 for each arm , which is shown in fig .
the hamiltonian of such quantum network reads where @xmath173 and @xmath174 are particle operators at site @xmath17 of chain @xmath175 and site @xmath176 of the arm @xmath133 .
they can be boson or fermion operators .
the conclusion for such model is available for the dynamics of a single flip in the analogue of spin network .
similarly as shown in ref .
@xcite , we can perform the following transformations@xmath177where @xmath178 and @xmath179 are also corresponding boson or fermion operators .
under the transformations , hamiltonian ( [ quantum network ] ) can be written as which indicate that the original quantum network , @xmath121 beam splitter can be decomposed into @xmath0 independent chains , one of them is the length of @xmath182 and the rest are all the length of @xmath171 .
based on the fact that a moving wave packet can propagate freely along the @xmath0 independent chains , we have the following processes through a straightforward algebra , we obtain a set of expressions @xmath184then for a given incident wave packet along any branch of a @xmath121 beam splitter , the probability amplitudes of all the output wave packets can be obtained exactly . | we propose a simple scheme to establish entanglement among stationary qubits based on the mechanism of resonance scattering between them and a single - spin - flip wave packet in designed spin network .
it is found that through the natural dynamical evolution of an incident single - spin - flip wave packet in a spin network and the subsequent measurement of the output single - spin - flip wave packet , multipartite entangled states among @xmath0 stationary qubits , greenberger - horne - zeilinger ( ghz ) and w states can be generated with success probabilities @xmath1 and @xmath2 respectively , where @xmath3 is the transmission amplitude of the near - resonance scattering . |
control problems for wheeled robots are usually described by nonholonomic systems .
a nonholonomic system arises when the dimension of configuration space is greater than the dimension of control . here and below
, the configuration space represents possible positions of a wheeled robot , i.e. , for a car - like robot it can be expressed in the following way : @xmath0 where @xmath1 corresponds to the reference point of the robot on a plane and @xmath2 corresponds to the orientation of the robot . for nonholonomic systems , motion in some directions is infinitesimally prohibited , but it is locally and globally possible ( through complex maneuvers of the system ) .
there is a well - known problem about parking of a car which can not move in a direction perpendicular to the direction of motion of the wheels .
in fact , every driver has faced this problem .
let us consider the simplest case where a car should be parked on an empty parking lot ( a plane ) with no other cars or static obstacles .
it is the classical hypothesis of `` rolling without slipping '' that provides the kinematic model of the car .
the control problem for this model is stated as follows : given a system of differential equations , fixed initial and final positions of the mobile robot and restrictions on one or two dimensional control , one should find a control law and the corresponding trajectory satisfying these conditions .
note that in order to control a real robot , the kinematic model should be derived from a dynamic one .
from a driver s point of view , cars have two controls : the accelerator and the steering wheel .
the position of a rear - wheel drive car can be defined by a vector @xmath3 , where @xmath4 is the midpoint of the rear wheels and @xmath5 is the angle of the car orientation which coincides with the direction of the rear wheels .
the direction of the front wheels is not fixed and corresponds to the steering control .
it is possible that each pair of wheels can be reduced to one wheel only .
moreover , if we are not concerned with the direction of the front wheels , the control system for a car - like robot is equivalent to the system for a wheel or a skate @xcite : @xmath6 where @xmath7 are respectively linear and angular velocities as controls .
note that the mechanical constraint for the steering angle of the front wheels of a car should be reduced to the constraint on @xmath7 .
such a system looks like the kinematic model of a unicycle or a two - driving wheel mobile robot .
however , the dynamical models of these systems are not identical .
the main difference between kinematic models lies in the admissible control domains which should be obtained from dynamical models .
further in the text , a car is identified with a mobile robot .
the problem is to find controls @xmath8 and the corresponding trajectory @xmath9 satisfying system ( [ sys ] ) with boundary conditions : @xmath10 where @xmath11 .
since such a formulation of the problem has a continuum of solutions in the general case , a cost functional should be added and minimized in order to choose the optimal solution among the family . the fact that we can choose the coordinate system @xmath12 allows us to always associate the initial position with the origin : @xmath13 further
, the paper considers a number of known approaches to solve the described above problem , which admits different restrictions on the control .
markov is one of the first mathematicians to work on such a problem @xcite . in 1887
, he studied four problems with application to railroad practice .
the first problem can be formulated geometrically in the following way . given two points @xmath14 in a plane , a positive number @xmath15 and a direction @xmath16 , the problem is to find the shortest smooth path @xmath17 joining @xmath18 and @xmath19 s.t .
@xmath20 is the tangent direction of @xmath17 at the point @xmath18 and the curvature of @xmath17 is bounded everywhere by @xmath21 .
the solution consists of an arc of a circle with radius @xmath15 and a straight segment or of two arcs depending on the position of the point @xmath19 ( see fig . [
fig : skmark ] ) .
such a solution can be applied to parking problem ( [ sys]),([positions ] ) , provided that the car is moving with a constant speed equal to one and is unable to move on a circle with radius smaller than @xmath15 , i.e. , @xmath22 and the direction of the car is of no importance at the desired end state , i.e. , @xmath23 is not fixed .
since the car is moving with the unit velocity , the length minimization is equivalent to the time minimization : @xmath24 other problems from @xcite involve the condition about the direction at the point @xmath19 , but also have some additional constraints on the curvature and its derivative . in 1957 ,
l. dubins @xcite solved problem ( [ sys])([timemin ] ) with fixed @xmath23 using geometric methods .
the solution is reduced to selection among 6 variants of smooth connection of arcs and straight segments ( see fig . [
fig : dub ] ) .
this model corresponds to the time - optimal control problem for a car moving with a constant velocity and with bounds on angular velocity .
such a car is called the dubins car . in 1990 ,
j.a . reeds and l.a .
shepp @xcite studied the so - called model of reeds
shepp car which can move forwards and backwards .
this case gives the following constraints on controls : @xmath25 ) .
however , the number of variants of shortest paths increases significantly , the worst case providing 68 variants .
optimal trajectories could contain cusp points , i.e. , points where the movement vector of the car changes to the opposite ( see example in fig .
[ fig : rsh ] ) . as in dubins work
, this characterization is done without obstacles . soon ,
simpler solutions for both models were obtained in @xcite and in @xcite via pontryagin s maximum principle @xcite .
in addition , h.j .
sussmann and g. tang reduced the sufficient family to 46 canonical paths . the reeds
shepp car is small time controllable from everywhere , i.e. , the reachable set of admissible configurations for any small time @xmath26 contains a small neighborhood of an initial point ( see geometric proof in @xcite ) .
the dubins car is locally controllable , i.e. , it is possible to find a path for the initial and final configurations that are arbitrarily close , but not small time controllable from everywhere , so the length of such a path does not converge to zero .
optimal control for both cars is a piecewise constant function .
both models serve as classical examples of time - optimal problems .
however , another criterion for choosing an optimal solution which minimizes the amount of car maneuvers can be considered . a classic problem from mechanics with such a variational principle
is described below . in 1744 , l. euler considered a problem on stationary configurations of an elastic rod @xcite . given an elastic rod in a plane with fixed endpoints and tangents at the endpoints ; should find the possible profiles of the rod with given boundary conditions .
euler derived differential equations for stationary configurations of a rod and described their possible qualitative types .
these configurations are called euler s elasticae . in 1880 , l. saalschutz obtained explicit parametrization of the curves @xcite .
these curves have rich history including max born s work @xcite on stability of euler s elasticae ( see fig .
[ fig : euborn ] ) , for more details see @xcite .
they appear as solutions to several problems of sub - riemannian geometry @xcite : nilpotent sub - riemannian problem on the engel group @xcite , nilpotent sub - riemannian problem with the growth vector @xmath27 @xcite , the problem of rolling a sphere over a plane @xcite .
in other words , we have system ( [ sys ] ) with @xmath28 , boundary conditions ( [ positions ] ) and a cost functional of energy to be minimized @xmath29 one of the solutions to the parking problem of a car can be expressed via the shape of an elastic rod @xcite , provided the car is moving forwards with a constant linear speed and with no boundary on steering .
time minimization in this case has no solutions ( curves tend to straight segments but can not reach them for general boundary conditions ) .
note that time @xmath30 is fixed in ( [ inte ] ) and is equal to the length of the corresponding rod .
jacobi elliptic functions @xcite are used in the expression of optimal control @xcite for a car moving along an elastica .
the global structure of all solutions of the problem is described in a series of works @xcite , the software and algorithm for numerical computation of optimal elasticae are given in @xcite .
a tool for plotting and evaluating generic euler s elastica can be found in @xcite .
all models considered above have one - dimensional steering control @xmath31 .
the reeds
shepp car has an additional discrete control for direction of movement .
a natural extension to it is the reeds - sheep car with varying speed and two dimensional control .
one of such models arises in sub - riemannian geometry @xcite .
consider system ( [ sys ] ) with no bounds on control @xmath32 .
the problem is to find a curve connecting two points ( [ positions ] ) with a minimum sub - riemannian length @xmath33 optimal solutions for this problem were obtained by yu .
sachkov in @xcite via geometric control theory @xcite .
a more general length functional reflecting the different weights of controls can be minimized : @xmath34 the extension is easily implemented by changing the configuration variables @xmath35 .
the parameter @xmath36 corresponds to the choice of scale in plane @xmath12 .
this solution can be applied to a mobile robot with two parallel driving wheels , the acceleration of each being controlled by an independent motor .
such a robot is called a differential drive robot .
the distance between the wheels may define the value of parameter @xmath36 .
consider a general model of a mobile robot with a trailer ( see fig .
[ fig : trailer ] ) .
if a robot is able to move forward only then such a system can usually be reduced to system ( [ sys ] ) , since the position of the trailer is of no importance .
such systems can use dubins paths ( see .
[ secdub ] ) or planar elastica ( see .
[ secela ] ) for the path planning problem .
that being said , the complexity of the parking problem increases significantly for a car with a trailer moving backwards .
a detailed survey on the various control strategies for the backward motion of a mobile robot with trailers can be found in @xcite .
formalization of control laws for such a problem requires the study of a nonlinear four - dimensional differential system with two - dimensional control which defines the linear and angular velocity . such a differential system has the following form : @xmath37 where @xmath38 are coordinates of the position of the car defined in sec .
[ car ] , @xmath39 is the angle of the trailer with respect to the car ; constant parameters @xmath40 define distances from the robot and the trailer to the connection point as shown in fig . [
fig : trailer ] .
the problem is to move a robot with a trailer from one configuration to another , i.e. , to find a path @xmath41 , s.t .
@xmath42 where @xmath43 such a problem has symmetries which translate solutions of the problem to other solutions .
the group of motions of the plane gives such symmetries and allows to move the initial position of a car to the origin ( [ origin ] ) .
another known symmetry is the similarity transformation @xmath44 , which changes configuration coordinates , constants and controls in the following way : @xmath45 this symmetry is used in the algorithm for trailer reparking , see subsec . [ reparking ] .
note that usually there is mechanical constraint @xmath46 .
such a constraint can be treated as an obstacle in @xmath47 .
the presence of obstacles can be described by inequalities and equalities for @xmath48 . in other words
we have a function @xmath49 which tells us if a point @xmath12 belongs to an obstacle .
the resulting constraint equations for configuration space have different forms for different shapes of a robot .
the simplest shape of a car - like robot is a circle because such constraint equations do not involve the angle parameter @xmath50 .
the algorithm for computing the configuration space obstacles , when the objects are polygons , is given in @xcite . in 1989 ,
j. barraquand and j .- c .
latombe @xcite , using standard results of differential geometry , nonlinear control theory and dynamic programming search , developed path planning algorithms for two robotic systems : car - like robots and trailer - like robots .
a car - like robot is kinematically similar to a rectangular - shaped car .
left fig .
[ fig : parking ] shows an example of maneuvering in an unstructured workspace represented as a @xmath51 bitmap with a maximal steering angle of the front wheels equal to 45 degrees ( the running time was about 2 minutes ) .
a trailer - car is kinematically similar to a vehicle towing a trailer .
right fig .
[ fig : parking ] shows an example with a simulated two - body trailer - like robot where the trailer has to maneuver in a cluttered workspace with a maximal steering angle equal to 45 degrees ( the running time was about 5 minutes ) .
their planners confirm controllability of such systems among obstacles in practice , but the solutions are not optimal . in 1994 ,
laumond , p. jacobs , m. tax and r. murray @xcite developed a fast and accurate planner for a car - like model in presence of obstacles , based on a recursive subdivision of a collision free path generated by a lower - level geometric planner that ignores the motion constraints .
the resultant trajectory is optimized to give a path that has near - minimal length in its homotopy class .
the model of a front - wheel - drive car with a constraint on the turning radius can be reduced to system ( [ sys ] ) with the constraints @xmath52 ) satisfying inequalities ( [ constraint3 ] ) are the same as for reeds
sheep s problem ( see the proof in @xcite ) .
the algorithm of the planner consists of three steps @xcite : * to plan a collision - free path for the geometric system , i.e. , without taking into account differential system ( [ sys ] ) . if such a path does not exist , neither does a feasible path .
the experimental results present two implementations based on two different geometric planners .
the first one runs for a polygonal robot ( see left fig . [
fig : laum94 ] ( a ) ) , the second one for a disk based on the voronoi diagram ( see right fig . [
fig : laum94 ] ( a ) ) . * to perform subdivision of the path until all endpoints are linked by a minimal - length path satisfying system ( [ sys ] ) and constraints ( [ constraint2 ] ) .
the shortest path for reeds - shepp car is used in order to link intermediate configurations along the path , generated in the previous step ( see fig .
[ fig : laum94 ] ( b ) ) . * to perform an `` optimization '' routine to remove extra maneuvers and reduce the length of the path ( see fig . [
fig : laum94 ] ( c ) ) .
the resulting paths are showed on fig .
[ fig : laum94 ] ( d ) . later in 1998 , j .-
laumond presented notes @xcite which describes a more general approach for the motion planning problem of a car and a car with one or two trailers .
he mentioned several steering methods for step ( b ) , including methods for nilpotent systems with piecewise and sinusoidal control . back in those days , steering with optimal control was possible only for car - like systems , so , the only possibility for general systems was to call on numerical methods . in 2013 , h. chitsaz @xcite considered the time - optimal control problem for system ( [ sysgen])([vf ] ) with @xmath53 and the constraints @xmath54 ^ 2 $ ] .
he revealed the structure of extremals through analytical integration of the system and adjoint equations .
it was shown that reeds - shepp curves are time - optimal for system ( [ sysgen])([vf ] ) , and that the general case corresponds to a planar elastica ( see subsec .
[ secela ] ) connecting an arc of a circle to a straight segment . a complete characterization of optimal trajectories and control synthesis for the problem is still an open question .
this paper considers a sub - riemannian problem for system ( [ sysgen])([vf ] ) , i.e. , the problem with boundary conditions ( [ positions2 ] ) and integral of sub - riemannian length ( [ intsub ] ) .
nowadays , computing optimal paths for such a problem is a challenging task .
sub - riemannian geometry considers optimal control problems with a system of linear in control differential equations and with the sub - riemannian length minimization . for two - dimensional control , such a system has form ( [ sysgen ] ) , where @xmath55 are smooth vector fields on manifold @xmath56 and @xmath57 .
the optimality criterion for two - dimensional control has form ( [ intsub ] ) .
system ( [ sysgen ] ) with ( [ positions ] ) and integral ( [ intsub ] ) describe an @xmath58-dimensional sub - riemannian problem with 2-dimensional control .
so far , only several sub - riemannian problems have been studied completely , mostly for @xmath59 .
the simplest problems are nilpotent and are defined by the simplest structure of lie algebras @xmath60 .
control system ( [ sysgen ] ) is called nilpotent if lie brackets of the corresponding vector fields @xmath55 vanish at a given length .
such systems provide a non - linear approximation for sub - riemannian problems .
linearization of control systems is usually used as a local approximation , but for system ( [ sysgen ] ) which is linear in control the linearization is a too rough approximation : since the dimension of control is less than the dimension of configuration space , the linearization can not be controllable .
nilpotent approximation preserves this important property of a control system .
the term `` nilpotent approximation '' is defined by a.a .
agrachev and a.v .
sarychev in @xcite , also by h. hermes in @xcite .
g. laferriere and h.j .
sussmann propose a method in the context of nilpotent systems in @xcite .
invariant construction of nilpotent approximation is proposed in @xcite . the exact solution to a control problem with a nilpotent system
gives an approximate solution to arbitrary system ( [ sysgen ] ) in a small neighborhood of the final point .
the method of nipotent approximation can be applied to control problem ( [ sysgen])([vf ] ) with arbitrary values of @xmath61 .
different solutions to the nilpotent problem give different approximations .
this article considers the solution in terms of optimal control ( [ intsub ] ) and uses a general algorithm , given by bellaiche in @xcite and further specified for system ( [ sysgen ] ) in @xcite , for constructing the nilpotent approximation .
it uses the maclaurin expansion of vector fields @xmath55 in privileged coordinates in the very same way as in @xcite for a mobile robot with two trailers .
the nipotent approximation for a mobile robot with a trailer is unique and has the following canonical form : @xmath62 problem ( [ syse])([vfe ] ) , ( [ intsub ] ) with arbitrary boundary conditions is called the nilpotent sub - riemannian problem on the engel group .
since the problem is invariant under left shifts on the engel group ( see the defenition in @xcite ) , we can assume that the initial point is the identity , i.e. , @xmath63 where @xmath64 .
the corresponding sub - riemannian problem has been studied quite recently in a series of works @xcite .
the problem of finding optimal synthesis in the general case @xmath65 is reduced to a system of three algebraic equations in elliptic functions and elliptic integrals .
it seems impossible to analytically solve such equations , therefore , a software for computing optimal trajectories for the nilpotent sub - riemannian problem on the engel group is being developed in wolfram mathematica @xcite .
this software has already been used to devise several algorithms for computing approximate paths close to optimal in terms of ( [ intsub ] ) for a mobile robot with a trailer ( see subsec .
[ park ] ) .
the algorithm proposed by a. bellaiche in @xcite is used to control a mobile robot with a trailer .
let us describe this algorithm applied to the problem ( [ sysgen])([positions2 ] ) , ( [ intsub ] ) .
commutators of the vector fields @xmath55 taken from ( [ vf ] ) have the following form : @xmath66 = { \left ( \begin{array}{c } \sin \theta \\
-\cos \theta \\ 0 \\ -\frac{l_r + l_t \cos \varphi}{l_t^2 } \end{array } \right ) } , \quad x_4 = [ x_1 , x_3 ] = { \left ( \begin{array}{c } 0 \\ 0 \\ 0 \\ -\frac{l_t + l_r \cos \varphi } { l_t^3 } \end{array } \right)}.\end{aligned}\ ] ] new coordinates @xmath67 are defined in the following way : @xmath68 where @xmath69 is the matrix composed from vectors @xmath70 and vector @xmath71 corresponds to the final point of the curve , see ( [ positions2 ] ) . with such changes , terminal point @xmath72 goes to the origin @xmath73 and initial point @xmath74 goes to point @xmath75 .
the next step is to find optimal control @xmath76 for problem ( [ sysgen ] ) , ( [ spacee ] ) , ( [ vfe ] ) with the boundary conditions @xmath77 and functional ( [ intsub ] ) .
it is possible to transform the boundary conditions to ( [ conden ] ) by inverting resulting controls and time on controls .
the controls @xmath78 should be applied to initial problem ( [ sysgen])([positions2 ] ) , ( [ intsub ] ) .
the corresponding curve @xmath79 can be found by numerical integration .
it gives an approximate solution to the problem close to optimal in terms of ( [ intsub ] ) . ]
let us consider the particular case of problem ( [ sysgen])([positions2]),([intsub ] ) : @xmath80 i.e. initial and final position of the car coincide one with each other , so the problem is to change the direction of the trailer . in this case
, according to ( [ gamma ] ) , the end point for the nilpotent problem on the engel group is the following : @xmath81 , @xmath82 for this case , there is an infinite number of optimal solutions which come to the same point @xmath83 in a figure - eight shape on the plane @xmath84 @xcite .
all of them can be used as the nilpotent approximation .
the corresponding optimal controls are described by jacobi elliptic functions : @xmath85 where , in this case , @xmath86 ( see @xcite ) and @xmath87 . on the other hand , using the parametrization of extremal trajectories @xcite , we get @xmath88{\frac{8 e(k_0)}{3 v_1}}. \label{sigma}\end{aligned}\ ] ] numerical experiments show that a trajectory of system ( [ sysgen])([vf ] ) starting from point @xmath89 with control ( [ cont1]),([cont2 ] ) and arbitrary @xmath90 always ends at point @xmath91 , where @xmath92 is a function which can be calculated numerically . for fixed @xmath93
, the parameter @xmath94 can be obtained from ( [ v]),([sigma ] ) .
a numerical algorithm is developed in order to find the value @xmath95 which satisfies @xmath96 .
see an example of reparking obtained by this algorithm with accuracy @xmath97 in fig .
[ fig : reparking1 ] . , @xmath98 the algorithm was tested for different @xmath99 and @xmath40 .
some cases provide rough accuracy @xmath100 , see an example in fig .
[ fig : reparking2 ] . usually , such accuracy arises for distant points .
one of the way to treat them is to use symmetry ( [ sym ] ) . for the reparking problem it is possible to consider controls for models with @xmath101 , @xmath102 and then translate them to controls for the model with @xmath61 .
such a transformation also changes the functional integral to ( [ intsub2 ] ) with @xmath103 .
therefore , using optimal solutions for the nilpotent sub - riemannian problem , it is possible to obtain an approximate solution for each weighted integral ( [ intsub2 ] ) with fixed @xmath36 .
this consideration is equivalent to choosing scale on the plane .
a similar trick has been used for sub - riemannian problem on @xmath104 regarding cusp avoidance in image analysis applications @xcite .
the example illustrated in fig .
[ fig : reparking2 ] is solved with such consideration , see fig . [
fig : reparking3 ] .
let us consider problem ( [ sysgen])([positions2]),([intsub ] ) with arbitrary @xmath105 .
the corresponding end point for the nilpotent problem on the engel group according to ( [ gamma ] ) is expressed in the following way : @xmath106 ] if @xmath65 , then there is a unique optimal trajectory for problem ( [ sysgen])([positions2]),([intsub ] ) @xcite .
note that for case @xmath107 it is always possible to consider an arbitrarily close solution corresponding to a point @xmath108 with small enough @xmath109 , which satisfies @xmath110 . in order to find optimal controls for the end point with @xmath65 , the corresponding system of algebraic equations in elliptic functions and elliptic integrals should be solved .
unfortunately , standard methods such as newton s method , secant method , random search and grid search do not solve the system , therefore a hybrid method should be used instead .
such a method is developed by combining standard methods , and it finds a solution in most of the cases .
a complete testing of the method is to be performed .
we have reasonable grounds to believe that it is possible to develop an approach which can solve the system for any @xmath83 .
thus , its detailed description is yet to be made in a future article . ] for arbitrarily close initial and final configurations the nilpotent approximation gives solution @xmath111 close to optimal .
however , such a solution usually does not connect the configurations .
starting from point @xmath112 it arrives at point @xmath113 .
if the distance @xmath114 is small enough , then the solution @xmath115 $ ] can be used to solve the parking problem .
otherwise , there should be found a way to improve the obtained solution .
[ fig : parking1 ] shows an example with @xmath116 . here and below , gray images of a robot and a trailer correspond to desired positions .
one of the ways to improve the approximation is to consider the parking problem with the constraints @xmath117 , where @xmath118 is an arbitrary point of curve @xmath119 .
this improvement can be repeated several times until the resulting curve comes to a point close enough to @xmath71 .
[ fig : parking2 ] shows an improved approximation of the example showed in fig .
[ fig : parking1 ] with @xmath120 .
black points correspond to points @xmath121 where the nilpotent approximation has been recalculated .
such an improvement is not always achieved with the desired precision , fig .
[ fig : parking3 ] illustrates an example with @xmath122 for improved approximation , where the initial approximation has precision @xmath123 . in order to obtain a more accurate solution ,
a specified algorithm should be developed on the base of the proposed approach .
more examples are shown in fig .
[ fig : parking4 ] with @xmath124 ( initial approximation has precision @xmath125 ) and in fig .
[ fig : parking5 ] with @xmath126 ( initial approximation has precision @xmath127 ) .
note that , as the last example shows , there is no restrictions on the parameter @xmath128 . ] ] ]
the article explores motion planning problems for a mobile robot with a trailer .
this is quite a difficult task even without obstacles .
the simplest case is given by the kinematic model of a car - like mobile robot , i.e. when the position of the trailer is not taken into account .
the article also includes a brief overview of the existing methods of solving a motion planning problem for a mobile robot and a mobile robot with a trailer .
one of them is based on the concept of nilpotent approximation .
different solutions , obtained from different classes of controls , are used to control a nilpotent system .
optimal control , in sense of minimization of controls , corresponds to nilpotent sub - riemannian problems .
recently , the nilpotent sub - riemannian problem on the engel group was solved .
this problem is given by a 4-dimensional control system with a 2-dimensional control and provides an approximate optimal solution to controlling a differential system for a mobile robot with a trailer .
the algorithm of nilpotent approximation is applied for solving the problem of reparking a trailer .
this algorithm is improved using a symmetry of the corresponding differential system for a mobile robot with a trailer . at the end of the article
a new algorithm for parking a mobile robot with a trailer is presented .
it will be applied for controlling a real mobile robot with a trailer .
dubins , l.e . , on curves of minimal length with a constraint on average curvature , and with prescribed initial and terminal positions and tangents , _ american journal of mathematics _ , 1957 .
79 , no . 3 , pp .
497516 .
pontryagin , l.s . ,
boltyanskii , v.r . ,
gamkrelidze , r.v . , and mishchenko , e.f . ,
_ the mathematical theory of optimal processes _ , ed .
: neustadt , l.w .
, translator : trirogoff , k.n .
, london : interscience , 1962 .
laumond , j .-
p . , feasible trajectories for mobile robots with kinematic and environment constraints , in _ proc . on intelligent autonomous systems _ ,
: hertzberger , l.o . and groen , f.c.a .
, new york : north holland , 1987 , pp .
346354 .
brockett , r.w . , and dai , l. , non - holonomic kinematics and the role of elliptic functions in constructive controllability , _ nonholonomic motion planning _
, eds . li , z. and canny , j. , boston : kluwer acad . publ . , 1993 , 121 .
agrachev , a.a . and
sarychev , a.v .
, filtrations of a lie algebra of vector fields and the nilpotent approximation of controllable systems , _ dokl . akad .
nauk sssr _ , 1987 ,
295 , no . 4 , pp .
777781 ( in russian ) ; translation in _
soviet math .
_ , 1988 , vol .
36 , no . 1 ,
104108 .
mashtakov , a.p . ,
algorithms and software solving a motion planning problem for nonholonomic five - dimensional control systems , _ programmnye sistemy : teoriya i prilozheniya _ , 2012 , vol . 3 , no . 1 , p. 329 ( in russian ) . | the work studies a number of approaches to solving motion planning problem for a mobile robot with a trailer
. different control models of car - like robots are considered from the differential - geometric point of view .
the same models can be also used for controlling a mobile robot with a trailer . however , in cases where the position of the trailer is of importance , i.e. , when it is moving backward , a more complex approach should be applied . at the end of the article , such an approach , based on recent works in sub - riemannian geometry ,
is described .
it is applied to the problem of reparking a trailer and implemented in the algorithm for parking a mobile robot with a trailer .
* keywords : * mobile robot , trailer , motion planning , sub - riemannian geometry , nilpotent approximation .
* msc 2010 : * 22e25 , 53a17 , 58e25 , 70q05 . |
the location of the sun in the rim of the loop i superbubble has been inferred from radio continuum data , kinematical data on the flow of local ism away from the center of loop i , data on gas - phase abundances in local ism , and the coincidence of the velocity of ism inside and outside of the heliosphere .
loop i is an evolved superbubble shell formed from stellar evolution in a subgroup of the sco - cen association , @xmath4 myrs ago ( e.g. * ? ? ? * ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
both the original dimensions found for the loop i bubble observed in 820 mhz @xcite , and more recent studies of ( * ? ? ?
* ; * ? ? ?
* h98a , h98b ) and @xcite , place the sun in or adjacent to the rim of a magnetic superbubble shell for an assumed spherical geometry .
the 1.4 ghz wolleben study defines two magnetic subshells of loop i , s1 and s2 , with magnetic pole directions differing by @xmath5 . comparisons between the radio continuum filaments of loop i and optical polarization data indicate that the radio filaments at distances of @xmath6 pc trace magnetic field lines , indicating that optical polarization is a suitable tracer of magnetic shells ( h98a ) .
both the kinematics and abundance pattern of local interstellar material ( lism ) suggest that the loop i remnant has expanded to the solar location @xcite .
lism abundances of the refractory elements mg , fe , and ca , show the characteristic enhancement indicative of grain destruction in interstellar shocks @xcite .
local interstellar gas , @xmath7 pc , and dust flow away from the center of loop i at a best - fit velocity of @xmath8 in the local standard of rest ( lsr , e.g. * ? ? ?
the first spectrum of backscattered ly@xmath9 emission from interstellar hydrogen inside of the heliosphere showed that the velocity of interstellar inside of the heliosphere is comparable to lism velocities @xcite .
together these data suggest that the magnetic field and spatial configuration of the lism can be used to test whether the loop i magnetic superbubble has expanded to the solar location .
the @xcite model of the s1 and s2 shells provides enough detail to make preliminary comparisons between lism data and the properties of these shells .
these comparisons provide interesting insights into the lism properties , and support the possibility that local ism within @xmath0 pc is dominated by the s1 and s2 shells .
superbubble expansion into ambient ism with equal magnetic and thermal pressures yields roughly spherical superbubbles during early expansions stages when magnetic pressure is weak compared to the ram pressure of the expanding gas @xcite , and bubbles elongated along the ismf during late stages of evolution @xcite .
the evolved shell is thicker near the ismf equatorial regions , where field strengths are larger due to flux freezing , than the polar regions of the shell where thermal pressure provides the main support for the shell . in media where magnetic pressure is weak , e.g. the ratio of thermal to magnetic pressure @xmath10 , the evolved bubble is more symmetric .
supernovae in sco - cen association subgroups have contributed to the evolution of the loop i superbubble during the past @xmath11 myrs .
the loop i superbubble ( and s1 , s2 ) expanded in a medium with a density gradient , because the initial supernova occurred in the molecular regions of the parent scorpius - centaurus association subgroups , while the subsequent bubble expansion occurred in the low density interior of the local bubble cavity @xcite . in this case
the external plasma @xmath12 may have varied irregularly across the expanding shell , so that the topology of the present day s1 and s2 shells may deviate from axial symmetry as well as sphericity .
the ismf direction at the heliosphere provides the most direct measure of whether the sun is embedded in the shell of the loop i superbubble .
several phenomena trace the field direction the weak polarization of light from nearby stars ( * ? ? ?
* ; * ? ? ?
* hereafter f07 ) , the flield direction in the s1 subshell of loop
i @xcite , the 3 khz emissions from the outer heliosheath detected by the two voyager satellites ( * ? ? ?
* f07 ) , the observed angular offset between interstellar and flowing into the heliosphere @xcite , and the 10 pc difference between the distances of the solar wind termination shock detected by the two voyager satellites ( e.g. * ? ? ?
the orientation of the plane midway between the hot and cold dipole moments of the cosmic microwave background is also within @xmath13 of the local ismf direction ( f07 ) .
this paper searches for evidence that the s1 and s2 shells affect the distribution of nearby ism within @xmath0 pc .
the topology of the s1 and s2 shells is discussed in
[ sec : shells ] .
section [ sec : mag ] shows that the direction of the ismf at the sun is consistent with the ismf direction in the s1 shell , similar to the location of the mid - plane between the cosmic microwave dipole moments , and consistent with the ismf direction inferred from heliosphere models .
the distribution of the ism in the s1 and s2 shells are compared to column densities towards nearby stars behind the shells ( [ sec : feii ] ) .
a similar comparison is made between the data and the s1 and s2 shells , however column densities appear instead to trace the strength of the local far ultraviolet ( uv ) diffuse radiation field ( [ sec : caii ] ) .
an appendix outlines the ionization equilibrium of .
@xcite has fit two separate spherical magnetic shells ( s1 and s2 ) to the low frequency ( 1.4 ghz and 23 ghz ) polarized radio continuum , which must have a relatively local origin because of the @xmath14 dependence of faraday rotation .
the ismf is assumed to be entrained in the expanding superbubble shell , with no deviation from spherical symmetry .
the sun is located in the rim of the s1 shell , which is centered @xmath15 pc away at galactic coordinates @xmath16 .
the upwind direction of the flow of local ism past the sun is within @xmath17 of of the s1 shell center .
the inner and outer radii of the s1 shell are @xmath18 and @xmath19 pc respectively .
wolleben described the s1 magnetic field direction by two angles , the angle between the field direction and the ngp @xmath20 , and the rotation about the ngp @xmath21 .
the s2 shell center is more distant ( @xmath22 pc ) and centered at higher galactic latitudes ( @xmath23 ) than the s1 shell , with an ismf direction near the north galactic pole . an alternate single - shell model for loop
i is also based on the shell and centers the feature at , @xmath24 ( h98a ) . as a first approximation of shell structure
, they are assumed spherically symmetric , although the ismf is seen in filaments interacting with denser clouds for more distant regions of loop i ( h98a ) a more detailed model of these evolved bubbles requires understanding the magnetic pressure .
the parameters provided for the s1 and s2 shells by wolleben are detailed enough for comparison with observations of the lism .
a three - dimensional ( 3d ) spherically symmetric model of the s1 and s2 shells is created for comparison with the local interstellar magnetic field and distribution of interstellar and .
the 3d configuration is initially constructed in the frame of each shell such that the north pole is at the local zenith .
the shell rims are filled with a uniform density of points .
the shell model is then rotated to the galactic coordinate system by and , and translated to the shell center in the galactic coordinate system .
the result is a model where a path through the shell measures the shell column density normalized to an arbitrary value , and the shell column density in any sightline varies according to the end point of the path .
[ fig : slices ] , right , shows the ism distribution in the s1 and s2 shells for slices parallel to the galactic plane and for 10 pc - wide intervals of z above and below the galactic plane .
[ fig : slices ] , left , shows sections of the two shells at distances of @xmath25 pc using an aitoff projection .
negative z - values are dominated by the s1 shell ( red ) , below z@xmath25 pc ( latitudes @xmath26 ) . at higher latitudes a given sightline may sample either , or both , of the shells .
the parameters of the 3d simulation place the sun in the rim of the s1 shell , but the uncertainties quoted by wolleben also allow the sun to be in the s2 shell . as additional data on nearby
ism become available , it should become possible to both constrain and test the s1/s2 models in more detail .
the s1 and s2 shells overlap in places .
one example is the sightline towards the stars @xmath9 oph ( hd 159561 ) , located 14.3 pc away at , @xmath27 , which has the strongest line observed towards any nearby star @xcite .
the s1 and s2 shells coincide at a distance of 12 pc in this sightline , suggesting that the line towards @xmath9 oph samples a region where the s2 shell collided with the s1 shell , possibly creating a shock so that recent grain destruction occurred . merging flows induce thermal instabilities that generate such filamentary structures @xcite , and the more distant gas in this sightline is also filamentary .
a second example of possible interacting shells is the nearby leo filament @xcite .
the orange symbol in fig .
[ fig : slices ] ( z=2535 pc ) shows the location of a tiny cold ( 20 k ) filamentary ( @xmath28 ) cloud in leo , located at , @xmath29 and at a distance of less than 42 pc @xcite .
if the cloud is at 40 pc , it is outside of both shells for the basic values for the shell distances and radii ( i.e. without invoking any uncertainties ) .
however if the cloud is nearby , or if the extreme values allowed by the uncertainties on the s1 and s2 shells are invoked , this filament may form where the two shells collide .
merging flows induce thermal instabilities that generate such filamentary structures as the leo filament @xcite , so the presence of this cold filament is consistent with the picture of the lism as dominated by two shell features .
polarization by charged irregularly shaped interstellar grains yields polarization vectors that are parallel to the ismf , because the induced magnetic torques naturally align the grains ( e.g. * ? ? ?
there is a patch of dust towards the fourth galactic quadrant ( @xmath30 ) , mainly in the southern hemisphere and in the upwind direction of the interstellar gas flowing through the heliosphere , where the ismf within 540 pc of the sun has been traced by very weak optical polarizations @xcite .
the tinbergen data were acquired in the southern hemisphere during 1974 , and northern hemisphere data during 1973 ( j. tinbergen , private communication ) .
tinbergen detected polarizations of @xmath31% , with @xmath32% , towards a few stars within 40 pc .
five of these stars level . ]
are close to the ecliptic plane , and offset by up to @xmath33 from the heliosphere nose towards positive ecliptic longitudes @xcite .
the mean position angles for the three stars near the nose with @xmath34 , are = @xmath35 in galactic coordinates , and = @xmath36 in ecliptic coordinates ( table 1 ) .
the position angle uncertainty is estimated by allowing q and u to vary over @xmath37 ( table [ tab : stars ] ) .
the nearest star towards the nose is 36 oph , at 6 pc , and it has a @xmath38 detection with @xmath39% . the polarization position angle of 36 oph ( = 19.9 , = 39.5 ) does not differ significantly from the mean position angle of the three more distant stars with @xmath40 polarizations in the nose region .
the polarizations of the tinbergen s sample , together with northern hemisphere data of @xcite , are plotted in fig .
[ fig : tinb ] .
an alternative catalog of nearby star polarizations is the comprehensive catalog assembled by @xcite ; however it is mainly based on measurements with larger uncertainties , and is therefore less useful for identifying very low polarization levels . in order to test the relation between the ismf in the s1 shell and the tinbergen polarization data ,
the ismf direction for the s1 shell was varied within the uncertainties on and to find the direction that is the most consistent with the optical data .
the best match to the optical polarization data was found for @xmath41 and @xmath42 .
the parts of the s1 shell within 30 pc are compared to the optical polarization data in fig .
[ fig : tinb ] , for galactic ( right ) and ecliptic ( left ) coordinates , for stars within 50 pc ; polarization vectors are plotted if @xmath43 . for the purpose of this figure , i use a center position for the s1 shell of ( , ) @xmath44 and radius 75 pc , so that the sun is located 3 pc outside of the shell rim .
the mean position angle of the s1 magnetic field ( for = 42 ) at the star locations is @xmath45 in ecliptic coordinates , which is within the uncertainties of the stellar polarizations .
a slightly larger value of exactly matches the mean position angles of the starlight polarizations , but violates the quoted uncertainties .
therefore , the ismf directions from the tinbergen data and s1 shell configuration are consistent to within the uncertainties .
the tinbergen stars with the strongest polarizations are @xmath46 from the pole of the s1 shell , which is consistent with the expectation of higher ismf field strengths where the shell expansion is perpendicular to the ismf direction . the best value for the s1 magnetic field direction close to the sun , derived from comparisons with these optical polarization data ( [ sec : mag ] ) ,
is @xmath41 and @xmath42 , corresponding to a local ismf direction towards , @xmath47 .
heliospheric asymmetries are caused by interactions with the interstellar magnetic field .
a widely used measure of the heliosphere distortion due to the ismf , which is inclined by the angle @xmath48 with respect to the ism flow vector , is the observed offset between the inflowing and directions @xcite .
the ismf also shifts the maximum emission originating in the outer heliosheath @xcite .
correcting the and directions to a common observation epoch yields an offset angle separation of @xmath49 between the two directions .
( the upwind direction of the flow in j2000 coordinates is @xmath50 , @xmath51 , witte , private communication ) .
these directions define a position angle , which is @xmath52 ( @xmath53 ) in ecliptic ( galactic ) coordinates , respectively .
large uncertainties are quoted because the upwind direction of the flow through the heliosphere is not precisely defined due to the @xmath54% of the interstellar lost to filtration in the hydrogen wall , the balance between radiation pressure and gravity affecting trajectories of atoms surviving to the heliosphere interior , and the production of secondary atoms inside of the heliosphere ( e.g. , * ? ? ?
* ) . for comparison , at the heliosphere nose location , , @xmath55 , the s1 shell with
= 42 gives a position angle = @xmath56 .
the position angle formed by the offset between and flowing through the heliosphere is marginally consistent with the s1 shell direction at the heliosphere nose .
the 10 au difference in the termination shock distance found by voyagers 1 and 2 , in 2004 and 2007 respectively @xcite , must be combined with the -offset to provide a more reliable constraint on models of the direction of the interstellar magnetic field affecting the heliosphere ( e.g. * ? ? ?
* ; * ? ? ?
* ) . and galactic latitudes in the range of @xmath57 .
the models also predict an ismf direction that is directed towards negative ecliptic latitudes . ]
the great circle that is midway between the hot and cold poles of the cosmic microwave background ( cmb ) dipole passes within @xmath58 of the interstellar upwind direction , and bifurcates the heliosphere nose @xcite .
the point of closest approach to the nose is at , @xmath59 , where the position angle of the cmb dipole mid - plane in ecliptic coordinates is @xmath60 ( uncertainties in the upwind direction , used to define the nose position , are included in this uncertainty , table 1 ) . for comparison , the s1 shell magnetic field direction at this location , for @xmath42 ,
corresponds to a position angle ( ecliptic coordinates ) of = @xmath61 .
this fact is mentioned here because the low-@xmath62 multipole moments of the cmb show symmetries related to the ecliptic geometry ( e.g. * ? ? ?
* ) , so that the symmetry of the cmb dipole moment around the heliosphere nose should also be of interest .
the upwind direction is @xmath58 above the ecliptic plane , so any coincidence between the cmb multipole moments and the ecliptic geometry is tantamount to a coincidence with the heliosphere morphology and/or to the interstellar magnetic field that shapes the heliosphere .
these ecliptic signatures on the cmb are not understood , but it is not unreasonable to postulate they arise from processes related to the local interstellar magnetic field and its effect on the heliosphere .
if the distribution of nearby ism is determined by the s1 and s2 magnetic superbubbles , then the s1 and s2 shell morphologies should be imprinted on the strengths of interstellar absorption lines .
[ fig : slices ] shows two views of the ism associated with the s1 and s2 shells .
interstellar absorption lines towards stars within 55 pc show that most of the the lism is warm , @xmath63 k @xcite with low average spatial densities , @xmath64 .
exceptions are the tiny dense clouds occasionally seen in and absorption @xcite .
models of the radiative transfer properties of the circumheliospheric ism show a partially ionized , low density cloud , @xmath65 , @xmath66 , and with temperature @xmath67 k determined from interstellar inside of the heliosphere ( * ? ? ?
* ; * ? ? ?
therefore , a suitable tracer of the s1 and s2 shell morphologies should be abundant , insensitive to cloud ionization , and undepleted .
there are no available data sets that meet all three requirements , therefore the criteria that the element be undepleted is dropped .
the best element for this study is then , which has been measured towards @xmath68 stars within 56 pc @xcite .
iron is predominantly singly ionized in the cloud around the heliosphere , with neutral and fe@xmath69 together containing less than 3% of the fe atoms ( sf08 ) .
a more difficult aspect of using to trace absolute ism densities is the factor of @xmath70 difference in the gas - phase abundances between dense cold and warm tenuous clouds , due to dust grain destruction by interstellar shocks including in local regions ( e.g. * ? ? ?
the alternative common element that traces both neutral and ionized gas is , however it has similar abundance variations as and the h and k lines may be more saturated than the lines .
therefore , i use column densities to trace the distribution of local ism .
three nearby stars are omitted from this discussion because they have known debris disks ( hd 215789 , hd 209952 , hd 216956 , * ? ? ?
comparisons between log and the column density of shell gas ( s1 and s2 ) in front of each star , versus a comparison between and the star distance , provide useful insights ( fig .
[ fig : length ] ) .
the shell column density towards each star represents the sum of the densities through the parts of the s1 and s2 shells foreground to the star , normalized to an arbitrary value .
this pathlength was constructed by assuming a column width of @xmath71 pc in order to smooth out uncertainties in the intrinsic shell parameters ( table 1 in * ? ? ?
because of this smoothing , stars within 6 pc are omitted from fig .
[ fig : length ] .
the bar at the bottom of the figures shows the column density range for stars within 6 pc . for low column densities , log @xmath72
, column densities tend to increase with star distance . for higher column density sightlines , log @xmath73 , clusters more tightly around the shell pathlength ( e.g. column density ) than around the star distance .
the exceptions to both comparisons are the stars hd 120315 ( a low column density , high - latitude star that should sample a long pathlength through the shells ) and hd 80007 ( a high column density low - latitude star , with a path that is tangential to the s1 shell ) .
the components in a sightline are summed together for this comparison .
the dashed lines in fig .
[ fig : length ] show a linear fit between the column densities and the ordinate .
the individual stars are listed in fig .
[ fig : panels ] , where the relative column densities of the s1 and s2 shells towards each star are shown .
a second property of the distribution of in these figures is that stars with galactic longitudes @xmath74 tend to have larger column densities , at a given distance , than stars in the opposite hemisphere .
this effect is not seen in or .
one possible reason is that the fe abundances differ between the two hemispheres .
the problem with this explanation is that it requires abundance variations over spatial scales of several parsecs , in relatively low velocity ism ( @xmath75 lsr ) , with the variations ordered by the arbitrary coordinate of galactic longitude .
an alternative explanation is that the lines for @xmath76 include cool unresolved clumps of ism .
the isrf towards @xmath77 is significantly larger than towards @xmath76 ( [ sec : caii ] ) , so that ism for @xmath76 experiences reduced heating because of the absorption of h - ionizing photons , which may allow unresolved ism clumps to coexist with warmer gas at the same velocity .
once the ionization of local ism is better understood , the morphology of the s1 and s2 shells can be adjusted to better represent the actual distribution of the local ism .
the stars within 6 pc show total column densities log @xmath78 . of these stars ,
the strongest lines are towards hd 155885 , hd 165341 , and hd 187642 in the upwind direction .
the distributions of s1 and s2 shell material towards each star are shown in fig .
[ fig : panels ] . stars with high ( low ) column densities are plotted as large ( small ) symbols for each direction .
the most evident property is that the s1 shell dominates the ism towards the southerly stars .
these can be used to predict whether a star ( or exoplanet system for example ) will be embedded in a cloud - like feature or in the local bubble plasma .
the assumed spherical morphology for s1 and s2 leads to the predictions that the white dwarf hd149499b ( wd1634 - 573 ) is embedded in in the relatively denser gas of the shell , while the white dwarf wd1615 - 154 will be embedded in the low density local bubble plasma .
interstellar is a recombinant species that traces the electron density as well as abundance variations . because is formed through recombination , it is a proxy for the electron density in nearby low density gas providing that abundance variations and the radiation field are understood .
if the isrf and electron densities are uniform throughout the s1 and s2 shells , then column densities would show a similar dependence on pathlength through the shells as seen for .
interstellar column densities are plotted against the pathlength through the s1 and s2 shells ( fig .
[ fig : lengthcaii ] , left ) and star distance ( fig .
[ fig : lengthcaii ] , right ) , using data from @xcite .
the column densities do not correlate with either the star distance or pathlength through the s1 and s2 shells .
sightlines where the column density is an upper limit are not included in this comparison .
higher column densities are found for stars with @xmath77 , as was seen for .
the mean column density is @xmath79% higher for stars in quadrants iii and iv , @xmath77 , compared to stars with @xmath76 .
the difference becomes a factor of two if the anomalously strong line towards @xmath9 oph is ignored . for @xmath76 , @xmath80@xmath81 ( 23 stars ) .
omitting @xmath9 oph , which has the strongest known for nearby stars ( e.g. * ? ? ?
* ) , gives a mean for the @xmath76 sample of @xmath80@xmath82 .
for @xmath77 , @xmath80@xmath83 ( 23 stars ) .
no similar effect is seen in ( or ) column densities towards nearby stars ( based on data in * ? ? ?
* ) . for stars within 50 pc
, the mean column density does not vary between the @xmath84 hemisphere and the @xmath74 hemisphere , and both show a mean value of @xmath85 for respective sample sizes of 25 and 27 stars .
a different picture emerges when ( and ) column densities are compared to the far uv radiation flux at the star .
the highest diffuse far uv fluxes are seen towards stars in the third and fourth galactic quadrants , @xmath77 , because of the low ism opacity in the local bubble interior , hot stars in the scorpius - centaurus association , and @xmath9 vir ( * ? ? ? * ; * ? ? ?
* go80,ow84 ) .
fluxes at 975 a are given for the 25 brightest stars , based on a survey by the @xmath86 satellite in the 9101050 a band and flux models ( ow84 ) .
at least 97% of the local flux at 975 a is provided by the stars with @xmath77 .
using the 975 a data for the 25 brightest stars ( ow84 ) , the 975 a flux was calculated at each star and is plotted against column densities ( fig .
[ fig : caii ] , left ) , and column densities ( fig .
[ fig : caii ] , right ) .
the and samples are different , with more than half of the stars within 20 pc , while most of the stars are beyond 20 pc .
all of the stars in the data set with 975 a flux levels larger than @xmath87 photons @xmath88
s@xmath89 a@xmath89 are in galactic quadrants iii and iv , and these stars tend to have larger column densities .
the isrf gradient would affect only through larger column densities of hii , since dominates in both ionized and neutral diffuse ism .
nearby h ii gas in regions with @xmath77 would explain variations , without violating the constraints ( which are determined from a similar star sample ) .
the dependence of column densities on the 975 flux is less simple because is a trace species formed by recombination , with an ionization potential of 11.87 ev versus 13.60 ev for . in the cloud around the sun ,
@xmath90 and /=63 ( sf08 ) . the isrf gradient near the sun ( fig .
[ fig : caii ] ) affects several ways .
higher radiation fluxes lead to higher electron densities , increasing @xmath91 recombination , and the overall fraction would increase .
higher radiation fluxes also increase the photoionization rate , but this effect does not appear to be dominant . in the local ism , photoionization appears to dominate over collisional ionization .
radiative transfer models of the ism surrounding the sun match available ionization data such as the / ratio ( sf08 ) .
low observed ar@xmath92 abundances towards nearby stars also indicate the dominance of photoionization @xcite .
the radiation flux at 1044 a capable of ionizing is traced by the 975 a radiation field , to within @xmath93% ( go80,ow84 ) , so that ionization rates should increase with the 975 flux .
the increase of with radiation flux is predicted by the photoionization equilibrium of ( see the appendix ) .
predicted values are plotted against the isrf for three different total h column densities , = + , in fig .
[ fig : caii ] . the observed increase of with higher fluxes is consistent with photoionization , and indicates that higher electron densities and h ii column densities are both significant factors in the line strengths .
the discussions in this paper are based on a search for evidence of the s1 and s2 shells in local ism data .
the s1 and s2 shells are assumed to be spherical and complete .
such simple assumptions are justified only in the initial stage of probing the ism distribution associated with a superbubble shell that has column densities too low for 21-cm measurements , and that has evolved into a very low density region of space .
several studies model the formation of the local bubble in terms of the energy injected into the ism by supernovae in the sco - centaurus association @xcite , but the connection between the loop i radio emission and very local ism has never been established .
the s1 and s2 shell models provide a basis for testing this connection .
early optical polarization data @xcite indicate that the ismf direction close to the sun agrees with the s1 shell ismf direction once the uncertainties in wolleben s angle ( 2007 ) are included . in principle
can be used to trace the ism distribution , since is arises in both neutral and ionized gas .
detailed comparisons between line strengths towards nearby stars and the projected pathlength through the s1 and s2 shells towards that star support , but do not prove , that the lines arise in shells . both the and data indicate that the portions of the s1 and s2 shells with @xmath77 will be more highly ionized than in the opposite hemisphere
. the data are not sufficient to distinguish an ionization gradient from an abundance gradient .
for this reason , the shells are better traced using ions with first ionization potentials less than 13.7 ev .
heating by radiation accounts for @xmath94% of the heating of the circumheliospheric ism , so shell regions exposed to the highest radiation flux should also be warmer ( an effect not explicitely included in the ca equilibrium discussion in the appendix )
. these models of the s1 and s2 shells assume spherically symmetric forms , which may be viable only for low density sightlines where magnetic and thermal pressures are comparable .
interstellar data are compared mainly to the s1 shell , which has the most favorable geometry for surrounding the sun according to @xcite .
the tinbergen data suggest a slow increase in polarizations with distance , and the position angles towards the nose are consistent with the optical polarization of more distant stars in loop i @xcite .
the most distant shell regions , in the galactic center hemisphere , have expanded into the high - extinction gas beyond @xmath95 pc that is associated with the sco - cen association ( see e.g. figs .
1,3 in @xcite or fig . 2 in @xcite ) and
show pronounced magnetic filaments ( h98a ) rather than a spherical shell geometry .
heiles points out that the synchrotron - emission ridges of loop i follow the distortion of the nearby global ismf , as traced by polarization data , and that loop i is not a shell for the high density regions .
the global ismf within several hundered parsecs is directed towards @xmath96 .
the expansion of the nearest portions of the s1 shell in a uniform field would yield an ismf close to the sun directed upwards with respect to the galactic plane .
@xcite model the properties of a magnetic superbubble @xmath97 myrs old and for the strong ismf case , where magnetic pressure dominates thermal pressure by a factor of @xmath98 .
the superbubble cavity is elongated in the direction parallel to the ismf , where the shell is thinner .
the shell is thicker and more extended in the radial direction , @xmath99 from the magnetic pole .
the region of strongest polarization for the tinbergen sample is @xmath99 from the magnetic pole .
non - uniform shell expansion , or a wrinkled shell , could explain the anomalous sightlines towards hd 120315 and hd 80007 ( see fig .
[ fig : length ] compared to fig .
[ fig : panels ] ) , while the tiny dense nearby leo cloud ( * ? ? ?
* lauroesch private communication ) , and the exceptionally strong line towards rasalhague ( hd 159561 , @xmath9 oph ) may indicate a region of merging shells .
the column densities for and are generally weaker for sightlines with @xmath100 , and stronger for stars with @xmath101 .
this effect may be either from the distribution of ionized gas , or abundance variations for fe and ca .
the effect is seen over small spatial scales of @xmath102 pc .
if the variation is due to abundance differences , then the ism close to the sun would have two different histories , although the flow velocities are similar .
the conclusions of this comparison between the s1 and s2 shells with lism markers can be briefly summarized : * the @xcite description of the loop i polarized radio continuum in terms of two shells , s1 and s2 , is viable and has sufficient detail to be tested against observational data .
for example , when the the s1 and s2 shell parameters are varied within the allowed uncertainty range , the nearby cold gas filaments in leo @xcite are seen to be produced where the two shells merge or collide ( [ sec : shells ] ) .
* the s1 shell magnetic field direction of @xcite , with = 42 , matchs the ismf direction derived from older polarization data @xcite of nearby stars near the ecliptic plane and heliosphere nose , but offset by up to @xmath103 from the heliosphere nose . the ismf direction implied by the s1 shell and polarization position angles together
is directed towards , @xmath47 . * for low column densities , log @xmath2
, the strength of the is better described by the star distance . for higher column densities ,
log @xmath3 , the strength of the is better described by the pathlength of the sightline through the s1 and s2 shells .
this result is based on a limited number of stars ( @xmath104 ) and requires confirmation using a larger data set ( [ sec : feii ] ) . *
the illumination of the s1 shell by the strong diffuse far ultraviolet interstellar radiation field in quadrants iii and iv , @xmath77 , explains the higher column densities observed for and in these galactic quadrants ( [ sec : feii ] , [ sec : caii ] ) .
an appendix evaluates the ionization equilibrium of features spaced around the shell , and shows that the local radiation field strength regulates the absorption line strengths .
* the ismf direction at the heliosphere nose is within @xmath105 of the angle of the great circle that is midway between the hot and cold hemispheres of the cmb dipole moment , and that also bifurcates the heliosphere nose . *
the s1/s2 shell model can be used to predict whether a star , or exoplanet system for example , is embedded in a cloud or in the local bubble plasma ( [ sec : feii ] ) .
the reverse is also true , that measurements of astrosphere properties will help constrain the distribution of ism associated with the shells .
* this scenario describing the influence of the magnetic superbubble s1 and s2 shells on the local ism , and as the origin of the interstellar magnetic field at the sun , is consistent with available data , but does not yet prove the s1/s2 model .
two kinds of data are required to substantiate this picture : ( 1 ) additional uv observations of tracers of both neutral and ionized interstellar gas , e.g. , , and features .
( 2 ) measurements of nearby weak interstellar polarizations at 0.01% levels or better .
the author would like to think nasa for research funding , in the form of grants nag5 - 13107 and nng05gd36 g to the university of chicago .
the column density can be determined from the assumption of photoionization equilibrium between and , for the radiation flux level @xmath106 at each star .
the highest fluxes of diffuse far uv radiation are seen towards the third and fourth galactic quadrants , @xmath77 .
some self - shielding of the ism in the two shells may occur , but the very low opacity of the local bubble interior suggests that the @xmath107 radial dependence of the isrf from the 25 brightest far uv stars is more germane for understanding local ism ionization , and the recombinant species that tracks the ionization .
for the temperature range considered here , @xmath108 k , collisional ionization is insignificant @xcite .
the equilibrium depends on the photoionization rate @xmath109 , the recombination rate from to @xmath110 , the electron density , the ca abundance @xmath111 , and the total hydrogen density @xmath112 : we define the ratio of the ionization and recombination rates at each star as @xmath115 for radiation flux @xmath106 , electron temperature @xmath116 , @xmath117 @xcite , and after parameterizing @xmath118 in terms of the properties of the circumheliospheric ism ( chism ) at the solar location based on model 26 in sf08 . for model 26 in sf08 ,
= 0.0654 ( and = 0.0554 ) , /=63.5 , @xmath119=6320 k , giving @xmath120 using eq .
[ eq : eq1b ] .
the ionization edge of is at 1044 a. for the ratio @xmath121 , @xmath106 is approximated by the total flux at each star @xmath122 from the combined distance - corrected radiation fields of the 25 brightest stars at 975 a ( based on fluxes as measured at the sun , with no opacity corrections , from * ? ? ?
the normalization factor @xmath123 is the 1044 a flux from @xcite as used for model 26 in sf08 , or @xmath124 photons @xmath88 s@xmath89 a@xmath89 . with this scaling , @xmath125 in eq .
[ eq : eq2 ] .
calcium abundances appear to vary by a factor of @xmath79 between cold and warm clouds @xcite .
i use the typical warm diffuse cloud calcium abundance of @xmath126 calcium atoms per hydrogen atom . then becomes : the observed relation between column densities and the 975 a flux seen in fig .
[ fig : caii ] is compared to predicted values of determined from eq .
[ eq : eq3 ] , for cloud temperatures in the range of 2,500 and 15,000 k and electron densities in the range of = 0.010.15 and positive detections of for stars within 55 pc ( using data from * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) , hd 87901 , hd 88955@xmath128 , hd 102124 , hd 103287 , hd 106591 , hd 106625@xmath128 , hd 108767@xmath128 , hd 112413 , hd 115892 , hd 120315 , hd 135742@xmath128 , hd 139006 , hd 141003 , hd 141378 , hd 148857 , hd 156164 , hd 159561 , hd 160613 , hd 161868 , hd 177724 , hd 177756 , hd 186882 , hd 187642 , hd 192696 , hd 203280 , hd 207098 , hd 209952 , hd 210418 , hd 212061 , hd 213558 , hd 215789 , hd 218045 , and hd 222439 .
the five stars marked with an asterisk have the highest 975 a radiation fluxes in this group . ] .
the missing parameter is the relation between the cloud temperature and electron density , @xmath116 in eq .
a3 , and for that i use the somewhat arbitrary relation @xmath129 in order to account for the increased cooling resulting from the collisional excitation of and fine - structure levels by electrons @xcite .
this assumption is required to fully specify the ionization equilibrium .
the lines labeled 20.00 , 19.5 , and 19.0 in fig . [ fig : caii ] show the predicted column densities for these assumptions and log = log @xmath130(+ ) = 20.00 , 19.5 , and 19.0 .
if the gas is clumpy , or conditions differ substantially from the chism gas , these estimates will break down . in the absence of a full 3d model of opacity over several hundred parsescs , more detailed comparisons between and the isrf require additional data on the cloud temperature , h density , or ionization .
however this comparison illustrates that reasonable assumptions for the parameters required to calculate the ionization equilibrium of yield reasonable predictions for the sensitivity of to the far uv radiation field .
, p. c. , dorschner , j. m. , geiss , j. , greenberg , j. m. , grn , e. , landgraf , m. , hoppe , p. , jones , a. p. , krtschmer , w. , linde , t. j. , morfill , g. e. , reach , w. , slavin , j. d. , svestka , j. , witt , a. n. , & zank , g. p. 1999
, , 525 , 492 , h. o. , allegrini , f. , crew , g. b. , demajistre , r. , frisch , p. c. , fuselier , s. a. , gruntman , m. , janzen , p. , mccomas , d. j. , mbius , e. , randol , b. , reisenfeld , d. b. , roelof , e. c. , & schwadron , n. a. 2009 , science , 326 , 964 , d. a. , kurth , w. s. , cairns , i. h. , & mitchell , j. 2006 , in american institute of physics conference series , vol . 858 , physics of the inner heliosheath , ed . j. heerikhuisen , v. florinski , g. p. zank , & n. v. pogorelov , 129134 , j. w. , howk , j. c. , andr ' e , m. , moos , h. w. , oegerle , w. r. , oliveira , c. , sembach , k. r. , chayer , p. , linsky , j. l. , wood , b. e. , ferlet , r. , h ' ebrard , g. , lemoine , m. , vidal - madjar , a. , & sonneborn , g. 2002 , , 140 , 19 , d. j. , allegrini , f. , bochsler , p. , bzowski , m. , collier , m. , fahr , h. , fichtner , h. , frisch , p. , funsten , h. o. , fuselier , s. a. , gloeckler , g. , gruntman , m. , izmodenov , v. , knappenberger , p. , lee , m. , livi , s. , mitchell , d. , mbius , e. , moore , t. , pope , s. , reisenfeld , d. , roelof , e. , scherrer , j. , schwadron , n. , tyler , r. , wieser , m. , witte , m. , wurz , p. , & zank , g. 2009 , science express , nov . , 13 , r. , ben - jaffel , l. , & grygorczuk , j. 2008 , in astronomical society of the pacific conference series , vol . 385 , numerical modeling of space plasma flows , ed . n. v. pogorelov , e. audit , & g. p. zank , 189 , n. a. , bzowski , m. , crew , g. b. , gruntman , m. , fahr , h. , fichtner , h. , frisch , p. c. , funsten , h. o. , fuselier , s. , heerikhuisen , j. , izmodenov , v. , kucharek , h. , lee , m. , livadiotis , g. , mccomas , d. j. , moebius , e. , moore , t. , mukherjee , j. , pogorelov , n. v. , prested , c. , reisenfeld , d. , roelof , e. , & zank , g. p. 2009 , science , 326 , 966 lccc position angles for
3 upwind stars with @xmath131 & @xmath132 & @xmath133 + mean s1 shell @xmath134-field towards 3 upwind stars & @xmath135 & @xmath136 + - offset at heliosphere nose ( , @xmath137 & @xmath138 & @xmath139 + s1 shell @xmath140field orientation at heliosphere nose & @xmath141 & @xmath142 + direction of cmb dipole midplane at ( , ) @xmath143 & @xmath144 & @xmath145 + s1 shell @xmath140field orientation at ( , ) @xmath146 & @xmath147 & @xmath148 + | many studies of the loop i magnetic superbubble place the sun at the edges of the bubble .
one recent study models the polarized radio continuum of loop i as two magnetic shells with the sun embedded in the rim of the s1 shell .
if the sun is in such a shell , it should be apparent in both the local interstellar magnetic field and the distribution of nearby interstellar material .
the properties of these subshells are compared to the interstellar magnetic field ( ismf ) and the distribution of interstellar and within @xmath0 pc of the sun .
although the results are not conclusive , the ismf direction obtained from polarized stars within @xmath1 pc is consistent with the ismf direction of the s1 shell .
the distribution of nearby interstellar with log @xmath2 is described equally well by a uniform distribution or an origin in spherical shell - like features .
higher column densities of ( log @xmath3 ) tend to be better described by the pathlength of the sightline through the s1 and s2 subshells .
column densities of the recombinant ion are found to increase with the strength of the interstellar radiation field , rather than with star distance or total pathlength through the two magnetic subshells .
the ion can not be used to trace the distribution of local interstellar gas unless the spatial variations in the radiation field are included in the calculation of the ionization balance , in addition to possible abundance variations .
the result is that a model of loop i as composed of two spherical magnetic subshells remains a viable description of the distribution of nearby low density ism , but is not yet proven . |
direct photon production is widely recognized as a process that is potentially important in determinations of the gluon distribution function . the next - to - leading - order ( nlo )
cross section for direct photon production has been given in refs .
the role of higher - order soft - gluon corrections has also been addressed more recently .
threshold resummation studies for direct photon production have appeared in refs .
@xcite while a joint threshold and transverse momentum resummation formalism has been given in ref .
@xcite . in a previous paper
@xcite we presented analytical and numerical results for the next - to - next - to - leading - order ( nnlo ) next - to - next - to - leading - logarithm ( nnll ) soft - gluon corrections for direct photon production .
here we increase the accuracy of our previous calculation by including additional subleading soft corrections .
our approach follows ref .
@xcite which in turn is based on and extends previous work on threshold resummation @xcite .
at lowest order , the parton - parton scattering subprocesses are @xmath1 and @xmath2 .
we define the mandelstam invariants @xmath3 , @xmath4 , and @xmath5 , which satisfy @xmath6 at threshold .
note that the photon transverse momentum is @xmath7 . here
we calculate the cross section @xmath8 in single - particle - inclusive kinematics in the @xmath9 scheme .
the soft corrections to the cross section appear in the form of plus distributions _
l(s_4)_+ with @xmath10 at @xmath11th order in @xmath12 beyond the leading order , while the virtual corrections appear in @xmath13 terms .
we begin with the nlo soft and virtual corrections in the @xmath14 scheme . a somewhat different notation from that used in ref .
@xcite has been adopted here , so the previously calculated terms are repeated here , as well .
the corrections to the parton - level cross section , @xmath15 , can be written for either subprocess as e _
= ^b_f_i f_j \{c_3^f_i f_j d_1(s_4 ) + c_2^f_i f_j d_0(s_4 ) + c_1^f_i f_j ( s_4 ) } , where @xmath16 is the renormalization scale , and the born terms are given by ^b_q|q = e_q^2 ( + ) , ^b_qg= - e_q^2 ( + ) , where @xmath17 is the charge of a quark of type @xmath18 , and @xmath19 with @xmath20 the number of colors .
also @xmath21 , @xmath22 , c_2^q|q=- -2c_f ( ) , c_2^qg =- c_f-(c_f+c_a ) ( ) , where @xmath23 is the factorization scale , @xmath24 , and @xmath25 , with @xmath26 the number of quark flavors .
we also define for use below @xmath27 and @xmath28 .
finally we write @xmath29 .
for @xmath30 we have c_1^=c_f ( ) + ( ) , and @xmath31\ln(p_t^2/s ) -(\beta_0/4)\ln(\mu_r^2/s ) + { c'}_1^{q \bar q}$ ] where @xmath32 is defined in eq .
( 3.11 ) of ref .
for @xmath33 we have c_1^= ( ) + ( ) and @xmath34 \ln(p_t^2/s)-(\beta_0/4)\ln(\mu_r^2/p_t^2 ) + { c'}_1^{qg}$ ] where @xmath35 is defined in eq .
( 3.8 ) of ref .
@xcite .
note that the nlo @xmath36 coefficients have also been presented in ref .
the notation for @xmath37 and @xmath38 is the same as before , while the notation for splitting @xmath39 into @xmath40 and @xmath41 terms for each subprocess is new and useful in presenting the nnlo expressions below .
next , we turn to the nnlo soft and virtual corrections in the @xmath14 scheme . these corrections can be written for either channel as e _ = ^b_f_i f_j ^(2)_f_i f_j .
[ nnlom ] for the @xmath42 process we have ^(2)_q|q&= & ( c_3^q|q)^2 d_3(s_4 ) + d_2(s_4 ) + & & + \{c_3^q|q c_1^q|q + ( c_2^q|q)^2 -_2 ( c_3^q|q)^2 - t_2 ^q|q + c_3^q|q ( ) + 2 c_f k . + & & .
+ c_a - } d_1(s_4 ) + & & + \{c_2^q|q c_1^q|q -_2 c_2^q|q c_3^q|q + _ 3 ( c_3^q|q)^2 - t_1^q|q + c_2^q|q ( ) + g^(2)_q |q . + & & + c_f + & & .
+ c_a - ( ) } d_0(s_4 ) + & & + r^q|qg ( s_4 ) .
[ nnloqqbar ] here @xmath43 , @xmath44 , and @xmath45 .
the function @xmath46 denotes a set of two - loop contributions @xcite and is given by g^(2)_q |q = c_f c_a ( _ 3 + _
2- ) + n_f c_f ( -_2 + ) .
we determine in the virtual corrections @xmath47 only the terms that involve the renormalization and factorization scales , denoted as @xmath48 and given explicitly by r^ q|qg&= & ^2 ( ) \{^2 - 2 _ 2 c_f^2 + _ 0 c_f+ c_f ( ) } + & & + ( ) ( ) c_f+^2 ( ) + & & + ( ) \{c_f^2 ^2 ( ) -c_f . + & & - _ 0 c_f -c_f ( ) + c_f^2(-11 _ 3 + _ 2 - ) + & & .
+ c_f c_a ( _ 3-_2 - ) + n_f c_f ( + ) } + & & + ( ) \{-c_f ( ) + t_1^q|q + ( ) + } , where @xmath49 and _ q / q^(2)=c_f^2(-_2 + _ 3 ) + c_f c_a(-_3+_2 + ) + n_f c_f ( -- )
. for the @xmath50 process we have ^(2)_q g&= & ( c_3^qg)^2 d_3(s_4 ) + d_2(s_4 ) + & & + \{c_3^qg c_1^qg + ( c_2^qg)^2 -_2 ( c_3^qg)^2 - t_2 ^qg + c_3^qg ( ) + ( c_f+c_a ) k .
+ c_f - _ 0 c_f } d_1(s_4 ) + & & + \{c_2^qg c_1^qg -_2 c_2^qg c_3^qg + _ 3 ( c_3^qg)^2 - t_1^qg + c_2^qg ( ) + g^(2)_qg .
+ & & + ( c_f+c_a ) + c_f k ( ) + c_a k ( ) + & & . + c_f - c_f _ 0 ( ) } d_0(s_4 ) + & & + r^qgq ( s_4 ) .
[ nnloqg ] the function @xmath51 denotes a set of two - loop contributions @xcite and is given by g^(2)_q g&=&c_f^2(-+_2 -_3)+ c_f c_a ( _ 3 -_2- ) + & & + c_a^2 ( _ 3 + _
2- ) + n_f c_f ( _ 2 + ) + n_f c_a ( -_2 - ) .
finally , the terms in @xmath52 that involve the renormalization and factorization scales , denoted as @xmath53 , are given explicitly by r^ qgq&=&^2 ( ) \{^2- ( c_f+c_a)^2 .
+ & & . + } + & & + ( ) ( ) + ^2 ( ) + & & + ( ) \{^2 ( ) .
+ & & - + & & --c_f^2 ( _ 3 + ) -c_a^2(_3 + ) + & & .
-c_f c_a ( _ 3+_2 + ) + n_f c_f ( + ) + n_f } + & & + ( ) \{- ( ) + t_1^qg + } , where _ g / g^(2)=c_a^2(+_3 ) -n_f(+ ) . for both processes the coefficients of the @xmath54 , @xmath55 , and @xmath56 terms were given previously in ref .
the additional subleading @xmath57 and @xmath13 terms presented here are new .
data from the ua-6 collaboration @xcite at @xmath58 gev.,height=4 ] data for the rapidity distribution from the ua-6 collaboration @xcite at @xmath58 gev.,height=4 ] data from the ua-6 collaboration @xcite at @xmath58 gev.,height=4 ] gev / c.,height=4 ]
in order to show the effect of including the new nnlo terms , the same procedure employed in ref . @xcite has been used .
first , a complete nlo calculation of the appropriate cross section is performed using a program @xcite which employs the phase - space slicing technique described in ref .
the original nlo calculation has been extended to include a complete nlo treatment of the bremsstrahlung contribution .
the set 2 fragmentation functions of @xcite have been used along with the cteq6 m parton distribution functions @xcite . in all cases the factorization and renormalization scales have been set equal to a common scale @xmath59 which has been chosen to be proportional to the photon s transverse momentum .
once the nlo results have been obtained , the approximate nnlo contributions can be added to them .
several examples are discussed below , comparing the nlo and nlo + approximate nnlo results . in fig .
[ fig1 ] the nlo ( dashed curves ) and nlo plus approximate nnlo results ( solid curves ) are compared to data @xcite from the ua-6 collaboration for proton proton interactions . in each case
the upper ( lower ) curve corresponds to the scale choice of @xmath60 @xmath61 .
the pattern demonstrated previously in fig .
@xcite is still found to be true , even after the addition of the newly calculated terms presented in this work .
the scale dependence of the nlo result is greatly decreased by the addition of the nnlo terms .
furthermore , one can see that for the scale choice of @xmath62 the nnlo scale dependent terms give a negligible contribution .
even with the nnlo contributions , the results lie somewhat below the data at the lower values of @xmath63 . in fig .
[ fig2 ] the rapidity dependence is shown for the ua-6 proton proton data .
again , the scale dependent nnlo terms give a negligible contribution for the choice @xmath62 and the overall scale dependence is greatly reduced when the nnlo terms are added .
as noted previously , the curves lie below the data over the majority of the rapidity range shown .
of course , this distribution is dominated by the contributions from the low end of the @xmath63 range , so this is no surprise , given the results shown in fig .
[ fig1 ] . in fig .
[ fig3 ] the photon @xmath63 distribution is shown for the case of @xmath64 interactions and compared to data from the ua-6 collaboration @xcite . whereas the @xmath65 reaction is dominated by the @xmath50 subprocess , the @xmath64 reaction receives additional significant contributions from the @xmath66 subprocess .
nevertheless , a pattern similar to that in the previous two figures is apparent here as well .
note , however , that for the case of @xmath67 , the nnlo contribution is negative , further reducing the scale dependence shown in fig .
[ fig3 ] . as for the @xmath65 case ,
the band formed by the theoretical curves lies somewhat below the data at the lower values of @xmath63 . finally , in fig .
[ fig4 ] a similar comparison is made to data from the e-706 collaboration @xcite .
the same behavior seen in the previous figures is evident here , although the theoretical band now lies significantly below the data at the lower end of the range covered by the data .
we have extended the results of ref .
@xcite to include additional nnlo subleading logarithms resulting from soft gluon corrections for direct photon production .
the additional terms are numerically small and the results remain qualitatively the same as in the previous analysis . in particular , the reduced scale dependence relative to nlo calculations remains .
the research of n.k . has been supported by a marie curie fellowship of the european community programme `` improving human research potential '' under contract number hpmf - ct-2001 - 01221 .
the research of j.o . is supported in part by the u.s .
department of energy . | previous work on soft - gluon resummation for direct photon production is extended to include additional subleading logarithmic terms through @xmath0 and some representative comparisons are made to experimental results from the e-706 and ua-6 collaborations .
the additional terms are small in magnitude , indicating good convergence properties to the level of accuracy calculated .
the scale dependence remains much smaller than that of the next - to - leading - order calculation . |
the formation of dark matter haloes has been studied extensively using numerical dark matter only simulations ( e.g. , * ? ? ? * ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
extending the insight from dark matter only simulations to include a theory of galaxy formation requires a method to link the formation of dark matter haloes to observable galaxy properties .
the most direct method available is hydrodynamical simulations , which model the co - evolution of dark matter and baryons ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) . by directly including hydrodynamics in structure formation simulations one can probe the thermal state and column density distribution of the intergalactic medium ( igm ) via the lyman-@xmath1 forest @xcite , the phase structure and heavy element composition of the circumgalactic medium ( cgm ) and igm @xcite , and the nature and rates of accretion of dark matter and baryons into galaxies @xcite .
since these simulations can follow the dynamics of both the dark matter and baryons down to small spatial scales , predictions can be made about their internal structure including the distribution of gas @xcite and the formation of stellar disks and bulges @xcite . as a result of this detailed information ,
simulations are a valuable tool for interpreting observational data and placing observed galaxies into a more complete evolution based cosmological context .
many important galaxy properties such as stellar mass have substantial uncertainties associated with their measurement ( e.g. , * ? ? ?
* ; * ? ? ?
* and references therein ) .
these uncertainties can be physical in origin ( e.g. , the slope of the imf , treatment of the thermally pulsating asymptotic giant branch phase , etc . ) or due to assumptions made during the spectral energy distribution ( sed ) fitting procedures @xcite . accurately estimating physical galaxy properties based on broadband photometric data points ( or even full seds ) is difficult because observed galaxy seds contain contributions from stars with a complex distribution of stellar ages and metallicities .
simplifying assumptions regarding the functional form of the star formation history , or a uniform stellar metallicity , are often assumed .
these assumptions can lead to systematic errors on the derived galaxy stellar mass , independent of physical uncertainties ( e.g. , * ? ? ?
similarly , derived galaxy structural properties such as bulge - to - disk decompositions or galaxy sizes are often determined inconsistently in simulations and observations @xcite .
a crucial difficulty when relating hydrodynamical simulations and observations is the translation of results between the two , a necessary step for accurate comparisons ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
there are two fundamental ways to make this translation , by either converting observed quantities into the space of physical parameters , or by generating ` mock ' ( or synthetic ) observations of the simulation output .
only the second approach provides a unique mapping , avoiding the uncertainties involved in a potentially non - unique inversion .
it also allows observational effects to be accounted for ( e.g. noise , survey geometry , ... ) , and enables observational tools and data analysis techniques to be run on the simulated observations as is . for comparison to observations of stellar light ,
this can be achieved in simulations by adopting stellar population synthesis ( sps ) models to assign light to all star particles within a galaxy based on their age and metallicity distributions @xcite .
these mock observations can be used to address potential issues in the way that sed fitting or photometric analysis is performed and even to directly fit galaxy seds rather than relying on sed modelling @xcite .
a large body of literature already exists within the idealized merger simulation community in this direction : @xcite used mock broadband images of idealized merger simulations @xcite to determine the conversion from observed galaxy pair counts into merger rates . @xcite and
@xcite used mock galaxy seds of idealized mergers to determine the lifetime and redshift dependent abundance of k+a galaxies , while @xcite used synthetic galaxy spectra to propose a new mid - ir diagnostic of agn . @xcite
investigated how well the true mass profiles of simulated massive compact galaxies could be recovered from their light profiles in different bands .
@xcite elucidated the physical meaning of the submillimetre galaxy selection , and @xcite tested how well the integrated ir luminosity traces the star formation rate .
similar work exists within the semi analytic community including the production of mock galaxy luminosity functions ( e.g. , * ? ? ?
* ; * ? ? ?
* ) and mock light - cone data @xcite based on dark matter only simulations @xcite .
the millennium run observatory ( mrobs , * ? ? ?
* ) built a theoretical virtual observatory capable of producing synthetic data products that could then be reduced and analyzed with observational software .
the mrobs synthetic observations covered forty filters where the light was assigned to each galaxy by combining information about the star formation history with stellar population synthesis models .
the resulting data products were formatted in a way that they could be analyzed using standard observational software packages @xcite and distributed using a web based relational sql database @xcite . in this paper , we describe the production of a mock galaxy image catalog based on the recent illustris simulation @xcite .
the illustris project has two basic principles : ( i ) build a comprehensive physical framework that allows galaxy populations to regulate their stellar mass growth appropriately and ( ii ) apply that physical framework in hydrodynamical simulations of large cosmological volumes . a detailed description of our adopted feedback module , which includes a model for star formation driven winds as well as agn driven outflows , was presented in @xcite and demonstrated to appropriately regulate galaxy populations across cosmic time in @xcite .
adopting this feedback model ensures that the galaxies within these simulations match a wide range of galaxy observable properties such as the redshift @xmath2 galaxy stellar mass function .
the main illustris simulation was run in a periodic box of size @xmath3 mpc with sufficiently high mass and spatial resolution ( i.e. @xmath4 and @xmath5 kpc ) to resolve a wide range of galaxy structures @xcite . using results from the illustris simulation ,
we employ stellar population synthesis ( sps ) templates to produce synthetic broadband galaxy images and integrated seds .
these mock images encode information about the spatial distribution and formation history of galactic stellar components .
the synthetic galaxy seds and images produced from the illustris simulation are unique compared to previous synthetic data catalogs .
compared against idealized isolated and merging galaxy simulations ( e.g. , * ? ? ?
* ; * ? ? ?
* ) , the illustris simulation galaxies are embedded in a proper cosmological context .
the galaxies in the illustris simulation accrete gas from the surrounding intergalactic medium , undergo major and minor mergers , and experience feedback as described in @xcite .
all of these processes are reflected in the stellar mass distribution and star formation history , which define the resulting mock images .
previous generations of synthetic images and spectra have been generated from
zoom - in " style simulations of individual ( or a small number of ) objects ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) . like the illustris simulation
, these simulations contain a proper cosmological context for galaxy growth , and they have there been used to study the growth of galactic disks and bulges .
our large - volume simulation approach is distinct from the `` zoom - in '' style simulations in that it allows us to analyze the properties of a large population of galaxies with varied formation histories and environments under uniform assumptions about the input physics ( e.g. , * ? ? ?
in contrast to semi - analytic based mock data products ( e.g. , * ? ? ?
* ) , our hydrodynamic approach allows us to self - consistently treat the internal galactic dynamics ( e.g. , bar formation , merger events , etc . )
that drive galaxy morphology evolution .
the resulting synthetic image catalog has many applications including building mass dependent average galaxy sed templates , isolating the origin of shelled galaxies , and understanding the viewing angle dependence of barred galaxies . in this paper ,
our main goal is to detail the methods for producing these images , explain the data format , and explain how the image catalog will be accessible .
we additionally demonstrate some initial science applications of the resulting synthetic images and spectra .
this paper is outlined as follows . in section [ sec : methods ] we describe the simulation that has been used to produce our mock image catalog as well as our methods for producing the mock galaxy image catalog based on its output . in section [ sec : data ] we describe the format for the mock galaxy images , present properties of the galaxy seds , and demonstrate multi - band image generation . in section
[ sec : example ] we provide a detailed application of this data , by determining stellar mass measurements using a popular sed fitting code to compare the sed derived stellar masses against the simulation based stellar mass determination . in section [ sec : discussion ] and section [ sec : conclusions ] we summarize and conclude .
.list of broadband filters used in the illustris virtual observatory , selected as a reasonably complete sampling of the historical , current and future bands used for the observation of galactic stellar light . [ cols="<,^,^ " , ] we store integrated global seds for each galaxy under the integratedquantities field .
example seds are shown in figure [ fig : gal_sed ] for one star - forming blue galaxy ( blue line , id=350781 ) , and one quiescent red galaxy ( red line , id=138415 ) .
both galaxies have stellar masses of @xmath6 .
the galaxies have star formation rates of @xmath7yr ( blue ) and @xmath8yr ( red ) .
while the blue galaxy still has a reservoir of star - forming gas , the red galaxy has quenched its star formation via agn feedback .
the two galaxies have noticeably different sed slopes with the star - forming galaxy having substantially more uv emission .
moreover , since a somewhat significant amount of light in the near ir comes from young stellar populations , an offset in the near ir sed can also be seen .
the fiducial seds do not take into account dust , which implies that the uv luminosities are overestimated due to missing sed attenuation .
this is demonstrated in figure [ fig : gal_sed ] , where the cyan and magenta lines show the star - forming galaxy s sed when applying the mappings hii region model and the cf00 empirical dust correction , respectively . in both cases , these models reduce the uv luminosity by approximating the obscuration of young stellar spectra . at optical wavelengths , the cf00 model depresses the sed by @xmath9 , while the mappings model is nearly coincident with the pure sb99 spectra .
however , since both galaxies are fairly massive and dominated by older stellar populations which contribute to the near ir light , the offset in the near ir portion of the sed is significantly less noticeable than in the uv . to present a more complete overview of the synthetic spectra properties ,
we show all 40,000 illustris galaxies by binning in narrow ( 0.01 dex ) stellar mass bins and calculate the median galaxy spectra at each wavelength as a function of galaxy stellar mass in figure [ fig : two_d_mean_spectra ] .
this figure shows the mean galaxy sed template that applies to our redshift @xmath0 synthetic galaxy spectra as a function of stellar mass .
the intensity of the median nir sed increases smoothly with increasing galaxy stellar mass .
little uv emission from both low- and high - mass galaxies can be identified .
the least massive galaxies produce little uv emission owing to their low star formation rates derived from their position on the star formation main sequence .
the high mass galaxies have low star formation rates owning to quenching from agn feedback in our model .
this view of the data also allows us to pick out several features of the mock spectra easily .
examples include the lyman limit at 912 , balmer break at 4000 , and h@xmath1 emission line at 6563 which are all included in the input sb99 template seds .
departures from the median trends identified in figure [ fig : two_d_mean_spectra ] can be substantial in the uv part of the spectrum owing to fluctuations in the star formation rates of galaxies even at similar stellar masses .
this variability is pronounced for high - mass systems ( i.e. @xmath10 ) where quenched and non - quenched galaxies sit in the same stellar mass bins . on the other hand ,
the variability of the median spectra are low for the most massive systems in the nir region owing to their nir luminosities being dominated by old stellar populations .
the median galaxy sed templates for the illustris galaxy population presented in figure [ fig : two_d_mean_spectra ] do somewhat depend on the adopted choice for the stellar population synthesis model .
we have compared the findings of figure [ fig : two_d_mean_spectra ] with the bc03-based outputs and found that qualitatively the mass binned median galaxy spectra generated using the bc03 stellar population synthesis models look very similar to those generated using the sb99 procedure .
the main distinction between these models can be seen in their uv spectra . whereas the sb99 models have little or no uv emission from old stellar populations , the bc03 catalogs contain non - negligible uv emission from old stars .
this causes the uv brightness of massive ( quenched ) galaxies to be substantially brighter for the bc03 model . in principle
, this difference also contributes to the uv emission of intermediate mass galaxies ( i.e. @xmath11 ) .
however , since the intermediate mass galaxies have substantial ongoing star formation dominating their uv spectra , any contribution to the uv spectra from old stellar populations is not easily identified .
spatially resolved galaxy images are stored in 36 bands , from which surface brightness maps and color images can be directly produced .
color galaxy images produced from our image pipeline were first presented in figure 13 of @xcite which shows 42 blue star forming galaxies and 42 red , passive galaxies as viewed through the sdss - g , -r , -i bands using the @xcite asinh scaling . here
we present additional applications of the broadband images by demonstrating : ( 1 ) the viewing angle dependence of a select set of galaxy types , ( 2 ) the visual appearance of a simulated galaxy in different filters , and ( 3 ) the time evolution of several elliptical galaxies exhibiting stellar shells .
figure [ fig : multi_angle_disks ] shows rgb color images of two disk galaxies made using the sdss - g , -r , -i bands as seen from three different directions to demonstrate the impact of viewing angle on perceived galactic morphology .
the left ( right ) most image is aligned ( anti - aligned ) with the angular momentum vector for the stars in the galaxy .
the central images uniformly transition between these two polar views .
we note that the full galaxy image catalog contains only four viewing angles for a galaxy , without a guarantee of any of the images being exactly ( anti-)aligned with the rotation axes .
the images presented here have been constructed after manually rotating the orientation of each galaxy to highlight the viewing angle dependence .
while the top galaxy ( id=283832 ) is relatively axisymmetric , the bottom galaxy has a prominent bar ( id=261085 ) .
the observable strength of the bar feature is a function of viewing angle , with the bar appearing strongest when viewed face - on and being completely unidentifiable when the galaxy is seen edge - on .
these mock images could then be used to derive quantitative correction factors for visually classified bar strengths , or to determine what fraction of strongly barred systems would have been classified as having weak bars , or no bars altogether , based on viewing angle .
similar examples of the impact of galaxy viewing angle can be constructed for merging / interacting galaxies ( e.g. , * ? ? ?
* ) , bulge - to - disk identification ( e.g. , * ? ? ?
* ) , and measured isophotal shape of elliptical systems ( e.g. , * ? ? ?
* ) . figure [ fig : bw_all_bands ] shows an example galaxy as seen through 12 different broadband filters , as labeled within the figure .
each image has been scaled independently using an asinh " scaling , with the non - linear transition set to the mean pixel intensity for that band .
bands increase in wavelength from top left to lower right ( effective wavelengths associated with the bands can be found in table [ table : broadbandfilters ] ) .
the uv bands reveal the emission from young star particles tracing the location of ongoing star formation in this system .
longer - wavelength bands trace stellar mass more efficiently and exhibit less band - to - band variation .
we note that this particular galaxy ( id=283832 ) was selected because it possesses a large stellar disk ( @xmath12 ) , with moderate levels of ongoing star formation , enabling us to make a clear demonstration of the wavelength - dependent image variation .
this is the largest and most massive disk galaxy in the simulation volume ; only a small fraction of the galaxies in the simulation are as well - resolved as the one presented here .
figure [ fig : shell_evolution ] shows a time series of rgb images from the rest frame sdss - g , -r , -i bands for three particular galaxies that we visually identify to possess stellar shells at redshift @xmath0 .
we trace these systems back in time using merger trees , identifying the most massive progenitor galaxy at each previous snapshot .
we generate images for a system at every snapshot using a fixed field of view of 120 physical kpc .
we find that merger events occur for all three systems within the past @xmath13gyr , which are then responsible for forming the observed present - day stellar shells .
this is consistent with the established theory of stellar shell formation via major @xcite and minor @xcite mergers . to identify the shell locations , we make monochromatic images of these systems in the johnson - k band ( as a good representation of the stellar mass ) and apply an unsharp mask procedure . in detail , we convolve the original image with a two dimensional gaussian with a standard deviation of 5 pixels .
we then subtract the gaussian blurred image from the original image , and discard all negative pixels to specifically highlight stellar light excesses relative to the blurred image .
the resulting masked image is shown in figure [ fig : unsharp_mask_shell_evolution ] using a logarithmic stretch . all high frequency image contributions
including noise from the discrete particle representation of stellar mass are amplified in the masked image . as a result
, the shells are more easily distinguished from the smoothly varying light component . using the masked images
, we identify the stellar shells down to fairly low galactocentric radii with @xmath133 - 5 shells being visually apparent in the systems at redshift z=0 .
stellar shells have been observed at low galactocentric radius using similar image masking techniques ( see * ? ? ? * for a detailed description ) .
we have confirmed the presence of low radii stellar shells by examination of a radial velocity versus radius phase diagram .
the presence of multiple systems with layered stellar shells in our cosmological simulation is of interest since the majority of previous theoretical work studying the evolution and characteristics of stellar shells formed via mergers has been done either with idealized merger simulations @xcite or with dark matter only cosmological simulations ( e.g. , * ? ? ? * ) .
since these systems are realized here within the full cosmological context , we can address some previously inaccessible questions , including : ( i ) the origin of very tightly bound shells , ( ii ) the observability lifetime of shells and other tidal features following merger events , and ( iii ) the predicted frequency of visible shells in the galaxy population . a deeper exploration into these topics using the illustris simulation and images presented here
is deferred to a separate , targeted study .
[ sec : stellarmass ] an examination of the distribution of galaxies of various morphological types as a function of mass , the morphology - environment relation , and the role of mergers in driving morphological evolution will be considered elsewhere ( snyder et al . , in prep . ) . here
we investigate a related problem which can be attacked with the image pipeline catalog : the accuracy of stellar mass estimates based on broadband photometric fitting .
the simple question we ask is : how well do broadband fitting routines recover the stellar mass from the simulations ?
similar topics have been addressed using semi - analytic models ( e.g. , * ? ? ?
* ) and idealized hydrodynamical simulations @xcite . for observational data , stellar masses are often determined by applying one of a few commonly adopted functional forms for the star formation history and finding the best - fitting stellar mass , galaxy age , star formation timescale , dust attenuation parameters , and metallicity to the observed sed .
one common assumption is an exponentially declining star formation history which is characterized by two free parameters : the galaxy formation time and the star formation timescale .
the simplest method for constraining these parameters is to perform a chi - squared minimization of the observed sed or broadband photometry relative to a grid of mock seds which uniformly cover the parameter space .
this is the general approach that is taken in , e.g. , fast @xcite . to perform this exercise
we assign light to a galaxy based on a stellar population synthesis template ( e.g. , the sb99 model ) , and then use the same ( or a similar ) stellar population synthesis model to find a best fit to the sed .
one might therefore worry that this is a circular argument , which will not shed any light on the validity of true " galaxy mass estimates .
however , the star formation history and heavy element composition for our simulated galaxies is known and properly reflected in the integrated galaxy sed , while the observational models must assume a functional form for the star formation history and metallicity .
the discrepancy leads , in fact , to the two recommendations for improved stellar mass determination , described below . to make the comparison as clear and idealized as possible
, we neglect dust in the simulated galaxy seds , and force fast to neglect dust obscuration as well ( i.e. set @xmath14 ) .
neglecting dust removes a large age / dust degeneracy which is present in real extragalactic data and mass measurements .
furthermore , we ignore dust re - radiation at long wavelengths which would add uncertainty into both our models and model sed fitting .
finally , we have not included any photometric and/or distance related errors in the galaxy s positioning .
we assume that we know a galaxy s distance exactly ( without error ) , and assign apparent magnitudes in each band appropriately .
owing to these simplifying assumptions , the comparisons that we perform below are highly idealized .
we expect that introducing any additional sed complications ( dust , distance errors , etc . ) would bring the derived stellar masses into worse agreement than what we present below because there would be additional characteristics of the intrinsic galaxy seds that are not fully accounted for by fast .
because of these simplifying assumptions , we will show below that we can obtain relatively good mass estimates for our mock galaxies based on their mock seds using only 5 bands .
however , we find that restricting the allowable age range and metallicity values that fast is permitted to sample can improve the derived stellar mass errors .
we employ the code fast @xcite to fit stellar population synthesis models to the artificially generated image pipeline broadband photometry .
we assume an exponentially declining star formation history ( @xmath15 ) to determine stellar masses based on the stellar - light - only galaxy photometry from the image pipeline .
for the initial comparison , we allow fast to select from the full range of input sed possibilities .
specifically , the galaxy age is taken in the range @xmath16 years with logarithmic steps of @xmath17 dex , the star formation timescale is chosen in the range @xmath18 years with logarithmic steps of @xmath19 dex , and the metallicity is selected from @xmath20 $ ] .
the mock galaxy photometry uses the sb99 templates while fast uses the bc03 templates .
we ignore dust attenuation in the mock photometry and restrict fast fits to neglect attenuation as well .
fast input files are generated by calculating rest - frame broadband fluxes for the five sdss bands ( u , g , r , i , z ) converted to units of flux density assuming the galaxies are at redshift @xmath21 using the same cosmology employed in the illustris simulation .
although fast has the ability to determine spectroscopic / photometric redshifts , in our initial comparison we supply correct redshifts to fast to avoid errors .
we compare the stellar mass derived from fast ( hereafter fast mass ) against the simulation determined stellar mass found in the subfind galaxy catalog ( hereafter subfind mass ) for all galaxies with integrated spectra in the image pipeline .
the left panel of figure [ fig : fast_masses ] shows a two dimensional histogram indicating the relationship between fast photometrically derived stellar mass and the subfind tabulated stellar mass for all 41,517 galaxies with more than 500 star particles at redshift @xmath0 .
the black line indicates a perfect 1:1 relationship ( i.e. the fast masses exactly equal the subfind masses ) . there is a tight relationship between these two quantities that nearly follows the ideal 1:1 scaling , with a slight offset and scatter . out of the 41,517 galaxies for which we perform this fitting , 1755 ( 4% ) disagree by more than a factor of two and 10 ( 0.02% ) disagree by more than a factor of five .
we calculated the average ( median ) star formation rate histories for galaxies in a mass bin ( @xmath22 < 10^{10}$ ] ) for several log(m@xmath23/ m@xmath24 ) values .
the resulting plot is shown in figure [ fig : delta_sfr_hist ] and can be used to partially determine what drives the fast mass estimate .
the median star formation histories evolve with this `` mass error '' parameter .
galaxies with the lowest mass error parameter , which correspond to galaxies whose masses have been most severely underestimated , have the oldest stellar populations and earliest peak in their sfr histories .
galaxies with old stellar populations will have the largest mass - to - light ratios , which drives their systematically underestimated stellar masses if fast does not choose a suitably large galaxy age . moving toward larger ( and positive ) mass error parameters
we find that the average sfr history becomes biased toward later times and the present day star formation rate monotonically increases .
for the most extreme positive mass error parameters , we find that they are heavily peaked toward very recent star formation .
the most extreme objects have either very early or late formation histories which are not captured in the fast best fit parameters .
the left panel of figure [ fig : fast_sub_ages ] shows the stellar half mass assembly time .
this can be compared against the center panel of figure [ fig : fast_sub_ages ] which shows the stellar half mass assembly time based on the assumed fast star formation history . in general the two distributions
look very different .
the fast age values span a larger range than found in subfind .
we plot the ratio of the subfind age to the fast age in the right panel of figure [ fig : fast_sub_ages ] as a function of galaxy mass .
there is significant scatter in this relation that is fairly mass dependent , with massive galaxies being assigned more accurate age estimates than low mass systems .
this mis - estimation of galaxy formation times can cause problems for the inferred mass - to - light ratios .
< 10^{10}$ ] are shown for galaxies .
the lines indicates the median star formation history for 50 galaxies with a mass error @xmath25m@xmath26= log(m@xmath23/ m@xmath24 ) closest to the prescribed value in the plot legend .
galaxies with underestimated stellar masses have systematically older stellar populations . , width=312 ] the right panel of figure [ fig : fast_masses ] shows the ratio of the fast masses to the subfind masses as a function of the subfind mass , which helps to identify some residual trends in the derived fast masses . there is a mass dependence in the average fast mass offset and scatter which can be ascertained from the distribution of galaxies outlined by the contour lines . we find that a principle origin of the mass dependent offset and scatter is the discrete and sparsely sampled metallicity values in the fast bc03 stellar population synthesis templates .
the standard bc03 templates used in fast draw from discrete metallicities values of @xmath20 $ ] without interpolation , and the result is a discrete jump in the derived stellar masses depending on which template metallicity is applied .
the left panel of figure [ fig : fast_metallicities ] shows the average ( mean ) template metallicity that fast applied to galaxies in each pixel .
since fast is forced to pick one of these sparsely populated metallicity values , the derived galaxy mass offsets correlate directly with different fast metallicity values .
the right panel of figure [ fig : fast_metallicities ] shows the mass weighted average stellar metallicity for these same galaxies .
we can gather from this plot that , even in this highly idealized experiment , fast is not able to accurately determine a galaxy s metallicity when using only 5 broadband filters . moreover , allowing for fast to assign metallicity values to each galaxy independently results in galaxies of similar intrinsic mass being assigned factor of @xmath27 different stellar mass estimates .
we have further checked and we discuss in section [ sec : fixed_metallicity ] that if we restrict fast to using only a single metallicity ( @xmath28 ) template , all derived galaxy masses fall along a single continuous relation , albeit with an offset from the desired @xmath29 relation .
here we re - compute fast masses by employing delayed or single - burst star formation histories .
both the delayed and single burst star formation histories are described by two parameters : a galaxy formation time and star formation timescale , @xmath30 . the galaxy formation time gives the lookback time when the galaxy started forming stars ( i.e. for all times before this , the star formation rate is assumed to be zero ) . in the case of the delayed star formation history ( @xmath31 ) ,
the star formation timescale modulates the decay timescale for the late time star formation rate decline . in the case of the single burst star formation history ,
the star formation rate is assumed to be constant after the galaxy formation time for one star formation timescale , after which it drops back to zero ( @xmath32 for @xmath33 ) .
figure [ fig : fast_alt_sfr ] shows the ratio of the fast mass to the subfind mass as a function of the subfind mass for the delayed star formation history ( left ) and single burst star formation history ( right ) .
the distribution of galaxies in the delayed star formation history space looks similar to what was found for the previously assumed exponential star formation history . by examining a number of best fit " star formation histories from the exponential and delayed models ,
we find that fast prefers very similar ( and young ) star formation history in many cases , explaining why the distribution of derived galaxy masses is so similar .
the distribution of derived galaxy masses for the single burst star formation rate history is more complex .
figure [ fig : fast_tru_t_and_z ] show the average metallicity ( left ) and age ( right ) .
there are three metallicities assigned to galaxies , causing mass estimate continuity issues as discussed above .
the metallicity alone can not explain the large spread in the galaxy population at @xmath34 .
the right panel of figure [ fig : fast_tru_t_and_z ] shows that there is a large age gradient , where the location along this gradient determines a galaxy s fast mass relative to the intrinsic subfind mass .
galaxies that are assigned old ages ( and long star formation timescales ) yield larger derived mass estimates ( owing to their larger mass - to - light ratios ) .
the origin of the large age spread for the low - mass galaxy population likely arises from the diversity of recent star formation histories of these low mass systems , which can substantially impact the best fit sed that fast selects .
it has been found , for instance , that when bursty systems are fit with single - component sfhs , their stellar masses can be significantly underestimated ( e.g. , * ? ? ?
* ; * ? ? ?
regardless of the underlying cause , we conclude that when fast assigns a very young age to a galaxy it is likely that the galaxy s mass will be underestimated . in the previous subsections
, we have found that allowing fast to pick a galaxy s metallicity during the sed fitting process can lead to mass - dependent errors in the derived galaxy masses .
here we re - compute fast masses while holding the assumed metallicity fixed at @xmath28 ( i.e. solar metallicity ) .
this is a commonly adopted assumption when deriving stellar masses for real observational data ( e.g. , * ? ? ?
the ratio of the fast mass to the subfind mass as a function of the subfind mass for a fixed metallicity with an assumed exponential star formation rate history is shown in the left panel of figure [ fig : fast_fixed_metallicity ] .
there is an offset between the derived mass and the intrinsic mass that evolves with intrinsic mass .
this trend is stronger than what was found when the metallicity was allowed to vary ( figure [ fig : fast_masses ] ) . at the low mass end ,
the systematic error in the estimated mass is @xmath35 dex , while the scatter about this value is roughly @xmath36 dex .
employing a fixed metallicity provides a more continuous and less broad distribution of mass estimates with respect to the intrinsic mass . in particular , holding the metallicity fixed in the fast calculation reduces some of the variation in the mass ratios previously found in section [ sec : fast ] .
however , the estimated mass error shows a clear mass dependent offset from the intrinsic mass and there is still substantial scatter in the derived masses .
the right hand panel of figure [ fig : fast_fixed_metallicity ] shows the average age that fast has assigned to each galaxy .
for all exercises up until this point , we have allowed the fast age selection to be selected from @xmath16 years .
examining fast s selection for the derived galaxy age value , there is a very clear residual trend where fast underestimates the mass of a galaxy when it assigns a young age ( i.e. @xmath37 years ) to that galaxy .
this is consistent with recent findings in the literature that young galaxy age assignment values can lead to underestimates in the stellar mass ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) .
in an attempt to improve the accuracy of the estimated masses , we restrict the galaxy age range from which fast can select to @xmath38 years .
the resulting ratio of the fast mass to the subfind mass as a function of the subfind mass ( still for a fixed metallicity with an assumed exponential star formation rate history ) is shown in figure [ fig : fast_age_restriction ] .
this case has the lowest systematic offset , mass dependence , and scatter in the estimated mass error of all of the cases presented in this paper .
the origin of this reduced bias is fairly clear : a relatively low mass of recently formed stars ( say , @xmath39 of a galaxy s total stellar mass ) can contribute substantially to the integrated sed especially at short wavelengths forcing fast to assume a relatively young age for galaxy as a whole when only a single exponential star formation rate history is assumed . by restricting the minimum galaxy age to be @xmath40 years
, we effectively prevent low galaxy ages ( with correspondingly low mass - to - light ratios ) from being assigned .
we would therefore encourage caution when allowing a single galaxy age parameter to vary in an unconstrained manner , when determining the stellar masses of real systems . for redshift @xmath0
we find that a lower age bound of @xmath41 years works well for our simulated galaxies .
however , this value will need to be lowered for high redshift galaxies .
it is also possible that this issue could be solved by adopting multi - component star formation histories ( e.g. , an exponential with superimposed bursts * ? ? ?
* ; * ? ? ?
* ) .
in this subsection we apply the fast broadband fitting procedure to spatially resolved galaxy images . by performing the fitting procedure in each pixel
we estimate the resolved surface density distribution which contains information about the dynamical state of the system as predicted by the simulation .
we first perform this analysis on the idealized image of a single redshift @xmath0 galaxy without applying any restrictions on the fast best fit parameters .
the central panel of figure [ fig : gal_1_fast ] shows the stellar mass surface density as derived from applying fast to each pixel of the image . for reference , the projected stellar surface density taken directly from the simulation for this galaxy is shown in the left panel of figure [ fig : gal_1_fast ] .
this galaxy ( id=283832 ) possesses a radially declining surface density profile as seen in the left panel of figure [ fig : gal_1_fast ] .
the stellar surface density profile is recovered in the fast analysis , with some detailed differences .
we find holes " in the fast stellar mass surface density profiles which are present in pixels which have significant contributions from young star particles .
these pixels are assigned young ages owing to the influence of the luminous young stellar population despite the fact that the majority of the mass in these areas is characterized by an older average age .
the low mass - to - light ratios associated with the young stellar ages result in an underestimation of the stellar mass surface density in these regions .
we next restrict the fast best fit parameters based on the minimum age and fixed metallicity criteria that we prescribe in the previous section .
the right panel of figure [ fig : gal_1_fast ] shows the stellar mass surface density map derived from fast when we restrict the minimum stellar age to @xmath40 years and fix the stellar metallicity to solar .
the derived stellar mass surface density map is qualitatively similar to that shown in the central panel of figure [ fig : gal_1_fast ] , with the surface density holes being somewhat less pronounced . because there are no pixels in this particular galaxy image which are mass dominated by young stellar populations we find that applying the minimum stellar age objectively improves the derived stellar mass surface density map . applying a fixed metallicity impacts the derived stellar mass surface density maps at the @xmath130.1 dex level , which is subdominant compared to the adopted age .
the illustris simulation observatory presented in this paper is a large mock galaxy image catalog constructed directly from a hydrodynamical simulation . in the past ,
large mock galaxy catalogs have been produced by applying semi - analytic models of galaxy formation to the output of dark matter only simulations ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
the approach used in the mrobs is to assign light to a bulge and disk component of a galaxy separately based on assumptions about their radial surface brightness profiles by using the bulge - to - total ratio , bulge half light radius , and disk scale height @xcite .
these galactic light components are then included in images by assuming a sky position angle and inclination .
after projecting all galaxies from a conic observational volume onto an image , point spread functions taken from the appropriate instruments are applied to yield mock light cones that compare against real observations favorably @xcite .
while these models have the advantage that they can be run on very large simulated volumes allowing them to produce mock light - cones and mock survey data they can only indirectly take into account the role that galactic dynamics ( e.g. , galaxy bar formation , major / minor mergers , disk instabilities , etc . ) play in morphological evolution .
in contrast , hydrodynamic simulations model gas processes and internal galactic dynamics self - consistently down to the gravitational softening limit . the resulting sample of simulated galaxies can therefore be directly converted into synthetic images which encode information about the stellar age distribution , stellar metallicity distribution , and stellar spatial distribution .
no further assumptions need to be made in order to generate a galaxy image .
the resulting light distribution from galaxies may therefore end up as disk - dominated , bulge - dominated , irregular , or highly asymmetric as dictated by the formation history and internal galactic dynamics of each system . in this paper
we have limited our focused to the construction of idealized images . in a companion paper , snyder et al . , ( in prep ) outlines our steps for adding image realism to the idealized galaxy images presented in this paper .
specifically , after creating the idealized images one can in post - processing rebin the images to an appropriate pixel scale , convolve with a point spread function , add background images or noise at a level consistent with a given instrument and exposure time .
analysis of the idealized images presented in this paper after being put through such post - processing steps will be soon presented in a companion paper ( snyder et al .
, in prep ) .
a series of important limitations need to be considered in order to properly exploit the illustris output
. an incomplete list of these is as follows :
the redshift @xmath0 gravitational softening length in the simulation is @xmath42 kpc , making small systems and small - scale features within large systems poorly resolved .
while we do observe barred galaxies formed during merger / flyby events , force softening can suppress disk instabilities and bar formation in non - interacting galaxies @xcite .
this can lead to a low fraction of barred systems , and potentially modulate the efficiency of stellar bulge formation ( to the extent that bars help drive bulge formation ) .
2 . the mass resolution of our models may impact results , particularly for low mass galaxies .
given our gas / star particle mass resolution of @xmath43 , young stellar populations will be stochastically sampled with a small number of particles .
this has been shown in the past to potentially bias non - parametric fitting parameters , such as the gini coefficient @xcite .
3 . our current models make use of a pressurized equation of state to handle the multi - phase ism in a sub - grid fashion @xcite .
features that are expected to exist in a true multiphase ism , such as spiral arm structure driven by pressure waves , will not be properly captured .
some simulated galaxies such as the disk galaxies presented in figure [ fig : multi_angle_disks ] do show signs of spiral features .
however , this spiral arm structure is driven by swing amplification @xcite of perturbations to the halo potential caused by the representation of dark matter by discrete simulation particles @xcite .
4 . our assumed physics ( and specifically our feedback implementation ) can influence galaxy morphology .
we find , for instance , a number of galaxies with enhanced rings of star formation in our simulation , which are likely caused by the interaction of star formation driven winds with the dense galactic ism .
the presence of these rings and/or the specific feedback formulation used in our model may influence visual and automated galaxy morphology classification or galaxy size growth .
our mock images and seds do not contain consider detailed radiation transfer .
the kilo parsec scale spatial resolution of the illustris simulation is slightly too coarse to carry out self - consistent dust modeling .
we have applied simple empirical corrections based on the cf00 model to the galaxy seds .
analysis of any systems spatial light distributions and/or sed properties where dust is expected to play a major role should therefore use caution . keeping in mind these caveats
, we still expect the illustris simulation observatory to be a useful and reliable resource in many situations , including : ( i ) addressing issues that can not be accessed with real data ( e.g. the viewing angle dependence of galaxy morphology and time evolution of galactic features ) , ( ii ) studying situations where the internal galaxy dynamics governing the spatial distribution of stars is of the essence ( e.g. in the formation and evolution of stellar shells or the viewing angle dependence of bar classification , as explored briefly in this paper ) , and ( iii ) where the full hydro simulation approach of the illustris simulation is likely robust ( e.g. for the formation of bars via mergers / interactions , the observability timescales for galaxy pairs and tidal features , the puffing up of massive galaxies via dry mergers , the formation of disk galaxies via in - situ star formation , etc . ) .
moreover , using an even - handed determination of galaxy morphology based on synthetic images will help us further identify the environments and mass ranges where our simulated galaxy populations are successfully or poorly replicating observations ( snyder et al .
, in prep ) .
we expect that the illustris simulation observatory will be an evolving resource , not only for the current illustris simulation , but also for next generation simulations with higher resolution , larger volume , and more sophisticated physics implementations .
the main goal of this paper is the presentation of a large synthetic galaxy image catalog based on a high - resolution full - volume cosmological hydrodynamical simulation .
we described our method for building this image catalog , in particular , the assignment of stellar light to each star particle within a galaxy .
we have made @xmath137,000 galaxy images in 36 broadbands plus integrated seds for @xmath1340,000 systems .
we have demonstrated that these broadband images can be used to generate high quality galaxy images .
our intention when constructing this mock galaxy image database is that it may serve as a useful tool for observers in the interpretation of extragalactic data . in principle
, the fits file format of our mock galaxy images will make it easy to process these images with existing image analysis tools , similar to what was done in @xcite .
this should facilitate galaxy selection from our catalog using selection criteria matched to observational targets .
it also implies that spatially resolved galaxy properties ( e.g. , half - light - radii , bulge - to - disk ratios , etc . ) can be derived using un - modified photometric image decomposition or fitting techniques . as a first application
, we showed that the sed fitting code fast is able to derive the `` correct '' galaxy mass within a factor of 2 for @xmath4495% of galaxies in our idealized sample .
we also uncovered some residual trends in the stellar mass estimates and argued that sparsely populated stellar population synthesis template metallicity values and low galaxy age assignments are responsible for the majority of the error in the photometrically derived galaxy mass values .
although our idealized comparison can not address systematic uncertainties such as the dependence of the mass - to - light ratio for a galaxy population on the shape of the imf , we have used our mock image pipeline to show that , overall , fast is able to derive reasonably accurate galaxy stellar masses for complex star formation histories using a relatively simple assumption for those histories .
we concluded that , for redshift @xmath0 , employing a lower age bound of @xmath41 years with a fixed solar metallicity produces the smallest errors in the stellar mass determination .
several additional applications of the mock image pipeline are currently under way .
we are undertaking automated non - parametric measurements of galaxy morphology for direct comparison with survey data ( snyder et al .
, in prep ) . among other goals , this will allow us to quantitatively assess the extent to which the morphological mix of the simulated galaxies match reality .
visual classification of the galaxy population for galaxies with @xmath45 stellar particles is also planned .
this selection criterion results in a sample of @xmath131000 galaxies with stellar masses @xmath46 , which we have verified contains a mix of both disks and elliptical systems .
visual classification is an alternative to the automated non - parametric classifications , and enables us to directly address whether , for example , the disk - to - elliptical fraction for simulated galaxies is comparable to the values measured from sdss in galaxyzoo @xcite . by making a fair comparison between cosmological simulations and large galaxy surveys , we can begin to connect theoretical ideas on the origins of galaxy structure with actual observations in ever more accurate ways .
the role that simulations play in shaping our understanding of galaxy formation will continue to increase as the models become increasingly more detailed and physically accurate . comparing the detailed structure and properties of individual galaxies within the context of their larger populations
is an important point of overlap between theory and observation , and mock catalogs based on realistic , synthetic observations of simulated systems are a fundamental requisite for maximizing the insight gained from that overlap .
we thank dave sanders and arjun dey for helpful suggestions on this work .
we thank patrik jonsson for his significant efforts to write , develop , and support the sunrise code .
gs acknowledges support from the hst grants program , number hst- ar-12856.01-a .
support for program # 12856 ( pi lotz ) was provided by nasa through a grant from the space telescope science institute , which is operated by the association of universities for research in astronomy , inc .
, under nasa contract nas 5 - 26555 .
cch is grateful to the klaus tschira foundation for financial support and acknowledges the hospitality of the aspen center for physics , which is supported by the national science foundation grant no .
vs acknowledges support by the dfg research centre sfb-881 the milky way system through project a1 , and by the european research council under erc - stg exagal-308037 .
lh acknowledges support from nasa grant nnx12ac67 g and nsf grant ast-1312095 . , g. , & springel , v. 2006 , in astronomical society of the pacific conference series , vol . 351 , astronomical data analysis software and systems xv , ed .
c. gabriel , c. arviset , d. ponz , & s. enrique , 212 | we present our methods for generating a catalog of 7,000 synthetic images and 40,000 integrated spectra of redshift @xmath0 galaxies from the illustris simulation .
the mock data products are produced by using stellar population synthesis models to assign spectral energy distributions ( sed ) to each star particle in the galaxies .
the resulting synthetic images and integrated seds therefore properly reflect the spatial distribution , stellar metallicity distribution , and star formation history of the galaxies . from the synthetic data products it is possible to produce monochromatic or color - composite images , perform sed fitting , classify morphology , determine galaxy structural properties , and evaluate the impacts of galaxy viewing angle .
the main contribution of this paper is to describe the production , format , and composition of the image catalog that makes up the illustris simulation obsevatory . as a demonstration of this resource ,
we derive galactic stellar mass estimates by applying the sed fitting code fast to the synthetic galaxy products , and compare the derived stellar masses against the true stellar masses from the simulation .
we find from this idealized experiment that systematic biases exist in the photometrically derived stellar mass values that can be reduced by using a fixed metallicity in conjunction with a minimum galaxy age restriction .
methods : numerical cosmology : theory cosmology : galaxy formation |
in this paper , we consider the @xmath0c+@xmath0c reaction as a case study to point out the problems for the inelastic scattering states which have so far remained unsolved , and to address particularly the magnitude problem for the inelastic scattering data .
theoretical calculations using the coupled - channels ( cc ) method fail to correctly predict the magnitude of the single-2@xmath1 and mutual-2@xmath1 states data together with the elastic scattering data . in order to get the magnitude right
, many futile theoretical attempts have been made for these states .
previous theoretical works show that the shapes of the central real potentials are actually correct , since they explain the elastic scattering data and predict the resonances at the correct energies with reasonable widths .
it appears that the failure of the standard methods is mostly related to the inelastic scattering data : the magnitude of the theoretical cross - sections is much smaller than the measured experimental data . in this paper
, we make further applications of a new coupling potential @xcite in describing the scattering observables of the @xmath0c+@xmath0c system .
for the spherical nuclei , the nuclear shape and also the shape of the potential between projectile and target nuclei are characterized by a constant radius @xmath2 , which defines the distance of the center of the nucleus from the surface . however , for a deformed nucleus , the radius parameter is no longer constant but depends on the angular location of the point ( @xmath3 ) .
the nucleus @xmath0c we study in this paper is strongly deformed and its collective excitation is taken into account by using the standard deformation procedure based on the taylor expansion .
if the interaction potential between two nuclei is taken to be @xmath4 , the taylor expansion about @xmath5 yields , @xmath6 here , the first term is the usual diagonal optical potential that describes only the elastic scattering and the second and third terms are used to describe the inelastic scattering and to obtain the coupling potentials for the single-2@xmath1 and mutual-2@xmath1 states .
@xmath7 in equation [ pot ] is given as : @xmath8 with @xmath9 as the projectile @xmath10 or the target @xmath11 .
@xmath12 is the deformation parameter and it is -0.6 for the @xmath0c nucleus . in the phenomenological analysis
, the real nuclear potential has the square of the woods - saxon shape , and the imaginary potential has the woods - saxon volume shape .
the parameters of the real and imaginary parts are taken from ref .
@xcite . for the microscopic analysis ,
the nucleon - nucleon double - folding potential @xcite is @xmath13 where @xmath14 and @xmath15 are the nuclear matter distributions for projectile and target nuclei respectively , and they are given by @xmath16\ ] ] where @xmath17=0.1644 @xmath18 , @xmath19=0.4988 @xmath20 and @xmath21=0.3741 @xmath20 for projectile and @xmath22=0.207 @xmath18 , c=2.1545 fm , and a=0.425 fm for target nuclei .
the m3y nucleon - nucleon effective interaction is taken in the form @xmath23 where @xmath24 .
the real and imaginary potentials are shown in figure [ comp ] and the parameters are given in table [ param ] labelled as df . in the new coupled - channels model
, we have replaced the usual first derivative coupling potential by a second - derivative coupling potential in woods - saxon form which is multiplied by the diffuseness parameter ( @xmath10 ) to normalize the units in the calculations .
the parameters are given in table [ param ] .
we have used both the phenomenological and microscopic potentials to analyze the experimental data of the @xmath0c+@xmath0c reaction at e@xmath25=74.2 mev , 93.8 mev and 126.7 mev .
the experimental data is taken from ref .
the results of our analyzes are displayed in figure [ ground ] for the ground , [ single ] for the single-2@xmath1 , and [ mutual ] for the mutual-2@xmath1 states in comparison with experimental data .
both double - folding and phenomenological potentials provide excellent agreement with the experimental data for the ground state at all energies and a good fit to the single-2@xmath1 state data .
however , the mutual-2@xmath1 state prediction is much smaller than the measured one : the standard cc model using double - folding or phenomenological potentials underestimates its magnitude by a factor of 3 to 10 as it has been previously observed @xcite .
varying the parameters and changing the shape of the real and imaginary potentials do not provide a complete solution to the problems of this reaction @xcite . in order to solve this problem
we have used a new coupling potential .
this potential has a second - derivative of woods - saxon shape and it is compared in figure [ comp ] with the standard coupling potential .
this new coupling potential has a repulsive part at short distances and an attractive part at large distances which is related to the orientation of two @xmath0c nuclei at short and large distances @xcite .
we have been able to obtain excellent agreement with all the available experimental data for the ground , single-2@xmath1 and mutual-2@xmath1 states by using this new coupling potential .
the parameters are shown in table [ param ] .
this new approach solves the magnitude problem of the mutual-2@xmath1 state data , which has been an outstanding problem with this reaction .
the results for the ground , single-2@xmath1 and mutual-2@xmath1 states are compared with the standard ones in figures [ ground ] , [ single ] and [ mutual ] .
in the present work , we have demonstrated that a consistent solution could be obtained for the problems of the @xmath0c+@xmath0c reaction over a wide energy range .
however , we achieve this by using a coupling potential which has a non - standard form . within the standard formalism ,
our findings using folding or phenomenological potentials are in agreement with the previous works @xcite .
although , within standard approach , these potentials give excellent agreement with the experimental data for the ground state , they are unable to provide a consistent solution to the problems of the inelastic scattering data .
this work therefore clearly shows that the standard deformation procedure based on taylor expansion is inadequate in describing such highly deformed nuclei .
it is obvious that the standard formalism should be questioned further .
this work is supported by the turkish science and research council ( tbitak ) : grant no : tbag-2398 and erciyes university - institute of science : grant no : fbt-04 - 15 and fbt-04 - 16 .
99 i. boztosun and w.d.m .
rae , phys .
c * 63 * ( 2001 ) 054607 .
m. el - azab farid and g. r. satchler , nucl .
a438 * ( 1985 ) 525 .
stokstad , r.m .
wieland , g.r .
c * 20 * ( 1979 ) 655 .
fry , dphil thesis , oxford university , 1997 .
rae , s.p.g .
chappell and p.e .
fry , nuovo - cimento * 110a * ( 1997 ) 1001 .
m. ito , y. sakuragi and y. hirabayashi , phys .
c * 63 * , 064303 ( 2001 ) .
r. wolf , o. tanimura , u. mosel and g.r .
satchler , z. phys . *
305a * ( 1982 ) 179 .
.the parameters of the real and imaginary potentials . for the new coupling potentials , the radius and diffuseness 0.7 fm and
the depth is 215.0 mev , 230 mev and 245 mev for @xmath26=74.2 mev , 93.8 mev and 126.7 mev . [
cols="<,>,^,^,^,>,>,>,>",options="header " , ] | we present the failure of the standard coupled - channels method in explaining the inelastic scattering together with other observables such as elastic scattering , excitation function and fusion data .
we use both microscopic double - folding and phenomenological deep potentials with shallow imaginary components .
we argue that the solution of the problems for the inelastic scattering data is not related to the central nuclear potential , but to the coupling potential between excited states .
we present that these problems can be addressed in a systematic way by using a different shape for the coupling potential instead of the usual one based on taylor expansion . |
the partonic structure of transversely polarized nucleons is one of their less known features ( for a review , see , e.g. , ref .
@xcite ) . the work presented here aims to contribute to the effort of shedding some light on it .
semi - inclusive deep inelastic scattering ( sidis ) is one of the proposed processes to access the parton distributions ( pds ) of transversely polarized hadrons .
sidis of unpolarized electrons off a transversely polarized target shows azimuthal asymmetries , the so called `` single spin asymmetries '' ( ssas ) @xcite .
the ssas are due to two physical mechanisms , whose contributions can be distinguished @xcite .
one of them is the collins mechanism , due to parton final state interactions ( fsi ) in the production of a hadron by a transversely polarized quark @xcite .
the other is the sivers mechanism @xcite , producing a term in the ssa which is given by the product of the unpolarized fragmentation function with the sivers pd , describing the number density of unpolarized quarks in a transversely polarized target .
the sivers function is a time - reversal odd , transverse momentum dependent ( tmd ) pd ; for this reason it was believed to vanish due to time reversal invariance . however , this argument was invalidated by a calculation in a spectator model @xcite , following the observation of the existence of leading - twist final state interactions ( fsi ) @xcite .
the current wisdom is that a non - vanishing sivers function is generated by the gauge link in the definition of tmds @xcite , whose contribution does not vanish in the light - cone gauge , as happens for the standard pd functions .
recently , the first data of sidis off transversely polarized targets have been published , for the proton @xcite and the deuteron @xcite .
it has been found that , while the sivers effect is sizable for the proton , it becomes negligible for the deuteron , so that apparently the neutron contribution cancels the proton one , showing a strong flavor dependence of the mechanism .
different parameterizations of the available sidis data have been published @xcite , still with large error bars .
new data , which will reduce the uncertainties on the extracted sivers function and will help discriminate between different theoretical predictions , will be available soon .
this experimental scenario motivates theoretical estimates of this quantity . since a calculation from first principles in qcd
is not yet possible , several model evaluations exist , in a quark - diquark model @xcite ; in the mit bag model @xcite ; in a light - cone model @xcite ; in a nuclear framework , relevant to proton - proton collisions @xcite . in here , a constituent quark model ( cqm ) calculation will be described @xcite .
cqm calculations of pds are based on a two steps procedure ( see , e.g. , ref .
first , the matrix element of the proper operator is evaluated using the wave functions of the model ; then , a low momentum scale is ascribed to the model calculation and qcd evolution is used to evolve the observable calculated in this low energy scale to the high momentum one , where dis experiments are carried out ( see , e.g. , ref @xcite ) .
such procedure has proven successful in describing the gross features of pds ( see , e.g. , @xcite ) and gpds ( see , e.g. @xcite ) , by using different cqms .
similar expectations motivate the present study of the sivers function .
the formalism of ref .
@xcite to calculate the valence quark contribution to the sivers function from any cqm is summarized here . a difference in the calculation of tmds , with respect to calculations of pds and gpds , is that the leading twist contribution to the fsi has to be evaluated .
all the technical steps of our procedure can be found in ref .
@xcite , where a workable formula is derived .
the main approximations have been : i ) only the valence quark sector is investigated ; ii ) the leading twist fsi are taken into account at leading order , i.e. only one - gluon - exchange ( oge ) fsi has been evaluated in the treatment of the gauge link ( see fig .
1 ) ; iii ) the resulting interaction has been obtained through a non - relativistic ( nr ) reduction of the relevant operator , according to the philosophy of constituent quark models @xcite , leading to a potential @xmath3 . in terms of the latter , the sivers function for the quark of flavor @xmath4 , @xmath5 , can be written ( cf .
fig . 1 for the labels of the momenta and helicities )
: @xmath6 where @xmath7 is the strong coupling constant , @xmath8 the proton mass , and @xmath9 eq . ( [ start2 ] ) , with @xmath10 given by eq .
( [ mu ] ) , provides us with a suitable formula to evaluate the sivers function , once the spin - flavor wave function of the proton in momentum space , i.e. the quantity @xmath11 , is available in a given constituent quark model .
a few remarks are in order .
first of all , the helicity conserving part of the global interaction does not contribute to the sivers function .
besides , one should notice that , in an extreme nr limit , the sivers function would turn out to be identically zero . in our approach
, it is precisely the interference of the small and large components in the four - spinors of the free quark states which leads to a non - vanishing sivers function , even from the component with @xmath12 of the target wave function .
effectively , these interference terms in the interaction are the ones that , in other approaches , arise due to the wave function ( see , e.g. , the mit bag model calculation @xcite ) .
there are many good reasons to use the isgur - karl model @xcite to test the performance of the approach .
first of all , the ik is the typical cqm , successful in reproducing the low - energy properties of the nucleon . in particular , in the ik model , one expects small corrections from terms @xmath13 . besides , one of the features of the ik model is that the oge mechanism @xcite , which reduces the degeneracy of the spectrum , is taken into account .
it is therefore natural to study our formalism , based on oge fsi , within the ik framework .
concerning pds , it has been shown that the ik model can describe their gross features , once qcd evolution of the proper matrix elements of the corresponding twist-2 operators is performed from the scale of the model to the experimental one @xcite .
reasonable predictions of gpds have also been obtained @xcite , and this makes particularly interesting the evaluation of the sivers function in the ik model .
the final expressions of the sivers function in the ik model are rather involved and not presented here , since they can be found in ref .
@xcite . to evaluate numerically eq .
( [ start2 ] ) , the strong coupling constant @xmath7 , and therefore @xmath14 , has to be fixed . here ,
the prescription introduced in the past for calculations of pdfs in quark models ( see , i.e. , ref .
@xcite ) will be used .
it consists in fixing the momentum scale of the model , the so - called hadronic scale @xmath15 , according to the amount of momentum carried by the valence quarks in the model . in the approach under scrutiny ,
only valence quarks contribute .
assuming that all the gluons and sea pairs in the proton are produced perturbatively according to nlo evolution equations , in order to have @xmath16 of the momentum carried by the valence quarks at a scale of 0.34 gev@xmath17 , as in typical low - energy parameterizations , one finds , that @xmath18 gev@xmath17 if @xmath19 gev .
this yields @xmath20 @xcite . for an easy presentation ,
the quantity which is usually shown for the results of calculations or for data of the sivers function is its first moment , defined as follows : @xmath21 , eq .
( [ momf ] ) , for the @xmath0 ( @xmath1 ) flavor .
the dashed curve is the result of the present approach at the hadronic scale @xmath15 .
the full curve represents the evolved distribution after standard nlo evolution ( see text ) .
the patterned area represents the @xmath22 range of the best fit of the hermes data proposed in ref .
, title="fig : " ] , eq .
( [ momf ] ) , for the @xmath0 ( @xmath1 ) flavor .
the dashed curve is the result of the present approach at the hadronic scale @xmath15 .
the full curve represents the evolved distribution after standard nlo evolution ( see text ) .
the patterned area represents the @xmath22 range of the best fit of the hermes data proposed in ref .
, title="fig : " ] the results of the present approach for the moments eq . ( [ momf ] )
are given by the dashed curves in fig .
2 for the flavors @xmath0 and @xmath1 .
they are compared with a parameterization of the hermes data , corresponding to an experimental scale of @xmath23 gev@xmath17 @xcite .
the patterned area represents the @xmath24 range of the best fit proposed in ref .
clearly , a different sign for the @xmath0 and @xmath1 flavor is found .
let us see now how the results of the calculation compare with the burkardt sum rule @xcite , which follows from general principles and must be satisfied at any scale . if the proton is polarized in the positive @xmath25 direction , in our case , where only valence quarks are present , the burkardt sum rule reads : @xmath26 where @xmath27 within our scheme , at the scale of the model , it is found @xmath28 mev , @xmath29 mev and , in order to have an estimate of the quality of the agreement of our results with the sum rule , we define the ratio @xmath30 obtaining @xmath31 , so that we can say that our calculation fulfills the burkardt sum rule to a precision of a few percent . the magnitude of the results is close to that of the data , although they have a different shape : the maximum ( minimum ) is predicted at larger values of @xmath32 .
one should anyway realize that one step of the analysis is still missing : the scale of the model , @xmath15 , is much lower than the one of the data , which is @xmath33 gev@xmath17 . for
a proper comparison , the qcd evolution from the model scale to the experimental one would be necessary .
unfortunately , the sivers function is a tmd pds and the evolution of this class of functions is , to a large extent , unknown . in order to have an indication of the effect of the evolution
, we perform a nlo evolution of the model results assuming , for the moments of the sivers function , the ones defined in eq .
( [ momf ] ) , the same anomalous dimensions of the unpolarized pdfs . as described in the previous section ,
the parameters of the evolution have been fixed in order to have a fraction @xmath34 of the momentum carried by the valence quarks at 0.34 gev@xmath17 , as in typical parameterizations of pdfs , starting from a scale of @xmath18 gev@xmath17 with only valence quarks .
the final result is given by the full curve in fig . 2 for the @xmath0 and @xmath1 flavor . as it is clearly seen
, the agreement with data improves dramatically and their trend is reasonably reproduced at least for @xmath35 .
of course a word of caution is in order : the performed evolution is not really correct . in any case , an indication of two very important things is obtained : i ) the evolution of the model result is necessary to estimate the quantities at the momentum scale of experiments , as it happens for standard pds @xcite ; ii ) after evolution , the present calculation could be consistent with data , at least with the present ones , still affected by large statistical and systematic errors .
one should notice that the agreement which is found is better than that found in other model calculations @xcite , especially for what concerns the fulfillment of the burkardt sum rule .
as we have discussed in the previous section , the experimental scenario which arises from the analysis of sidis off transversely polarized proton and deuteron targets @xcite is rather puzzling .
the data show a strong , unexpected flavor dependence in the azimuthal distribution of the produced pions . with the aim at extracting the neutron information to shed some light on the problem ,
a measurement of sidis off transversely polarized @xmath2he has been addressed @xcite , and two experiments , planned to measure azimuthal asymmetries in the production of leading @xmath36 from transversely polarized @xmath2he , are forth - coming at jlab @xcite . here , a recent , realistic analysis of sidis off transversely polarized @xmath2he @xcite is described .
the formal expressions of the collins and sivers contributions to the azimuthal single spin asymmetry ( ssa ) for the production of leading pions have been derived , in impulse approximation ( ia ) , including also the initial transverse momentum of the struck quark .
the final equations are rather involved and they are not reported here .
they can be found in @xcite .
the same quantities have been then evaluated in the kinematics of the planned jlab experiments .
wave functions @xcite obtained within the av18 interaction @xcite have been used for a realistic description of the nuclear dynamics , using overlap integrals evaluated in ref .
@xcite , and the nucleon structure has been described by proper parameterizations of data or suitable model calculations @xcite .
the crucial issue of extracting the neutron information from @xmath2he data will be now thoroughly discussed . as a matter of facts , a model independent procedure , based on the realistic evaluation of the proton and neutron polarizations in @xmath2he @xcite , called respectively @xmath37 and @xmath38 in the following , is widely used in inclusive dis to take into account effectively the momentum and energy distributions of the bound nucleons in @xmath2he .
it is found that the same extraction technique can be applied also in the kinematics of the proposed experiments , although fragmentation functions , not only parton distributions , are involved , as it can be seen in figs . 1 and 2 .
in these figures , the free neutron asymmetry used as a model in the calculation , given by a full line , is compared with two other quantities .
one is : @xmath39 where @xmath40 stands for `` collins '' or `` sivers '' , @xmath41 is the result of the full calculation , simulating data , and @xmath42 is the neutron dilution factor . the latter quantity is defined as follows , for a neutron @xmath43 ( proton @xmath44 ) in @xmath2he : @xmath45 and , depending on the standard parton distributions , @xmath46 , and fragmentation functions , @xmath47 , is experimentally known ( see @xcite for details ) .
@xmath48 is given by the dotted curve in the figures .
the third curve , the dashed one , is given by @xmath49 i.e. @xmath2he is treated as a nucleus where the effects of its complicated spin structure , leading to a depolarization of the bound neutron , together with the ones of fermi motion and binding , can be taken care of by parameterizing the nucleon effective polarizations , @xmath37 and @xmath38 .
production ( full ) in jlab kinematics , and the one extracted from the full calculation taking into account the proton effective polarization ( dashed ) , or neglecting it ( dotted ) .
the results are shown for @xmath50=0.45 and @xmath51 gev@xmath17 , typical values in the kinematics of the jlab experiments.,title="fig : " ] production ( full ) in jlab kinematics , and the one extracted from the full calculation taking into account the proton effective polarization ( dashed ) , or neglecting it ( dotted ) .
the results are shown for @xmath50=0.45 and @xmath51 gev@xmath17 , typical values in the kinematics of the jlab experiments.,title="fig : " ] one should realize that eq .
( [ extr-1 ] ) is the relation which should hold between the @xmath2he and the neutron ssas if there were no nuclear effects , i.e. the @xmath2he nucleus were a system of free nucleons in a pure @xmath52 wave .
in fact , eq . ( [ extr-1 ] ) can be obtained from eq .
( [ extr ] ) by imposing @xmath53 and @xmath54 .
it is clear from the figures that the difference between the full and dotted curves , showing the amount of nuclear effects , is sizable , being around 10 - 15 % for any experimentally relevant @xmath32 and @xmath50 , while the difference between the dashed and full curves reduces drastically to a few percent , showing that the extraction scheme eq .
( [ extr ] ) takes safely into account the spin structure of @xmath2he , together with fermi motion and binding effects .
this important result is due to the peculiar kinematics of the jlab experiments , which helps in two ways .
first of all , to favor pions from current fragmentation , @xmath50 has been chosen in the range @xmath55 , which means that only high - energy pions are observed .
secondly , the pions are detected in a narrow cone around the direction of the momentum transfer .
as it is explained in @xcite , this makes nuclear effects in the fragmentation functions rather small .
the leading nuclear effects are then the ones affecting the parton distributions , already found in inclusive dis , and can be taken into account in the usual way , i.e. , using eq .
( [ extr ] ) for the extraction of the neutron information . in the figures
, one should not take the shape and size of the asymmetries too seriously , being the obtained quantities strongly dependent on the models chosen for the unknown distributions @xcite .
one should instead consider the difference between the curves , a model independent feature which is the most relevant outcome of the present investigation .
the main conclusion is that eq .
( [ extr ] ) will be a valuable tool for the data analysis of the experiments @xcite .
the evaluation of possible effects beyond ia , such as final state interactions , and the inclusion in the scheme of more realistic models of the nucleon structure , able to predict reasonable figures for the experiments , are in progress .
it is a pleasure to thank alessandro drago and santiago noguera for interesting conversations .
this work is supported in part by the infn - cicyt agreement , by the generalitat valenciana under the contract ainv06/118 ; by the sixth framework program of the european commission under the contract no .
506078 ( i3 hadron physics ) ; by the mec ( spain ) under the contract fpa 2007 - 65748-c02 - 01 and the grants ap2005 - 5331 and pr2007 - 0048 .
v. barone , a. drago and p. g. ratcliffe , phys .
* 359 * ( 2002 ) 1 .
j. c. collins , nucl .
b * 396 * , 161 ( 1993 ) .
p. j. mulders and r. d. tangerman , nucl .
b * 461 * ( 1996 ) 197 [ erratum - ibid .
b * 484 * ( 1997 ) 538 ] .
a. m. kotzinian and p. j. mulders , phys .
b * 406 * ( 1997 ) 373 .
d. boer and p. j. mulders , phys .
d * 57 * ( 1998 ) 5780 .
a. bacchetta , m. diehl , k. goeke , a. metz , p. j. mulders and m. schlegel , jhep * 0702 * , 093 ( 2007 ) .
d. w. sivers , phys .
d * 41 * , 83 ( 1990 ) , phys . rev .
d * 43 * , 261 ( 1991 ) .
s. j. brodsky , d. s. hwang and i. schmidt , phys .
b * 530 * , 99 ( 2002 ) . s. j. brodsky , p. hoyer , n. marchal , s. peigne and f. sannino , phys . rev .
d * 65 * , 114025 ( 2002 ) .
j. c. collins , phys .
b * 536 * , 43 ( 2002 ) .
x. d. ji and f. yuan , phys .
b * 543 * , 66 ( 2002 ) .
a. v. belitsky , x. ji and f. yuan , nucl .
b * 656 * , 165 ( 2003 ) .
a. drago , phys .
d * 71 * , 057501 ( 2005 ) .
m. anselmino , m. boglione , u. dalesio , a. kotzinian , f. murgia and a. prokudin , phys .
d * 71 * , 074006 ( 2005 ) , m. anselmino , m. boglione , u. dalesio , a. kotzinian , f. murgia and a. prokudin , phys .
d * 72 * , 094007 ( 2005 ) .
[ erratum - ibid .
d * 72 * , 099903 ( 2005 ) ] a. v. efremov , k. goeke , s. menzel , a. metz and p. schweitzer , phys .
b * 612 * , 233 ( 2005 ) ; j. c. collins , a. v. efremov , k. goeke , s. menzel , a. metz and p. schweitzer , phys .
d * 73 * , 014021 ( 2006 ) .
a. de rujula , h. georgi and s. l. glashow , phys .
d * 12 * , 147 ( 1975 ) .
n. isgur and g. karl , phys .
d * 18 * , 4187 ( 1978 ) ; phys .
d * 19 * , 2653 ( 1979 ) .
[ erratum - ibid .
d * 23 * , 817 ( 1981 ) ] .
m. burkardt , phys .
d * 69 * ( 2004 ) 091501 ; phys .
d * 69 * ( 2004 ) 057501 .
s.j . brodsky and s. gardner , phys .
b 643 , 22 ( 2006 ) .
e-06 - 010 proposal to jlab - pac29 , j .-
chen and j .- c .
peng spokespersons ; e-06 - 011 proposal to jlab - pac29 , e. cisbani and h. gao spokespersons . s. scopetta , phys .
d * 75 * , 054005 ( 2007 ) .
a. kievsky , m. viviani , and s. rosati , nucl .
a 577 , 511 ( 1994 ) .
wiringa , v.g.j .
stocks , and r. schiavilla , phys .
c 51 , 38 ( 1995 ) .
a. kievsky , e. pace , g. salm , and m. viviani , phys .
c 56 , 64 ( 1997 ) ; e. pace , g. salm , s. scopetta , and a. kievsky , phys .
c 64 , 055203 ( 2001 ) .
d. amrath , a. bacchetta , and a. metz , phys . rev .
d 71 , 114018 ( 2005 ) . c. ciofi degli atti , s. scopetta , e. pace and g. salm , phys .
c48 , 968 ( 1993 ) . | a formalism is presented to evaluate the sivers function in constituent quark models .
a non - relativistic reduction of the scheme is performed and applied to the isgur - karl model .
the results obtained are consistent with a sizable sivers effect and the signs for the @xmath0 and @xmath1 flavor contributions turn out to be opposite .
the burkardt sum rule is fulfilled to a large extent . after the estimate of the qcd evolution of the results from the momentum scale of the model to the experimental one , a reasonable agreement with the available data
is obtained .
a calculation of nuclear effects in the extraction of neutron single spin asymmetries in semi - inclusive deep inelastic scattering off @xmath2he is also described . in the kinematics of forth - coming experiments at jlab
, it is found that the nuclear effects arising within an impulse approximation approach are under control .
address= dipartimento di fisica , universit degli studi di perugia , via a. pascoli , 06100 perugia , italy , altaddress= infn , sezione di perugia , via a. pascoli , 06100 perugia , italy address= departament de fisica terica , universitat de valncia + and institut de fisica corpuscular , consejo superior de investigaciones cientficas + 46100 burjassot ( valncia ) , spain address= dipartimento di fisica , universit degli studi di perugia , via a. pascoli , 06100 perugia , italy , address= departament de fisica terica , universitat de valncia + and institut de fisica corpuscular , consejo superior de investigaciones cientficas + 46100 burjassot ( valncia ) , spain , altaddress= th - division , ph department , cern , ch-1211 genve 23 , switzerland |
the physics of the high - temperature superconducting oxides is determined by the interplay between the charge and spin degrees of freedom , ultimately responsible for the superconductivity itself .
a variety of interesting phenomena exists already at low doping when the oxide layers are insulating . in la@xmath0sr@xmath1cuo@xmath2 ( lsco ) , the insulating ( spin - glass ) region corresponds to doping @xmath4 , with incommensurate magnetism which exists down to the boundary with the antiferromagnetic phase ( at @xmath5 ) , and even inside the nel region ( @xmath6 ) .
@xcite a popular point of view favors an explanation of the incommensurate magnetism based on the tendency of the holes to form stripes . " @xcite however , experimental data on variable range hopping ( vrh ) ( see the review ref . ) , unambiguously indicate localization of holes for @xmath4 and therefore support an approach based on a purely magnetic scenario , where a spiral distortion of the spin background is generated by localized holes
. the corresponding theory explains quantitatively the variety of magnetic and transport data in lsco .
@xcite magnetic phenomena in the low - doping region reflect , in addition to the heisenberg exchange , the presence of anisotropies in the spin - spin interactions , such as dzyaloshinsky - moriya ( dm ) and xy terms . in the present paper
we consider the nel phase , @xmath6 . in this phase
the anisotropies confine the spins to the @xmath7 plane and fix the direction of the nel vector to the @xmath8-orthorhombic axis .
moreover , the dm interaction induces a small out - of - plane spin component that is ferromagnetic in the plane ( weak ferromagnetism ) but staggered in the out - of - plane @xmath9-direction .
this component can be easily influenced by an external magnetic field applied in different directions , as it has been recently addressed both experimentally @xcite and theoretically .
@xcite for example , a perpendicular field ( @xmath10 ) can cause an alignment of the out - of - plane moments via a spin - flop transition at a critical field @xmath11 , determined by the competition between the dm and inter - layer heisenberg exchange ( typically @xmath12 ) .
@xcite perhaps most intriguingly , the in - plane resistivity ( along with the @xmath9-axis resistivity ) decreases by as much as @xmath13 across such a transition .
@xcite the magnitude of the magnetoresistance ( mr ) shows a rapid increase only below @xmath14 k @xcite where lsco exhibits vrh conduction .
@xcite this implies that the mr is accumulated mostly in transitions between localized states .
therefore it is very natural to assume that the large negative mr is due to an increase of the hole s localization length as it was suggested in the first experimental paper . @xcite from theoretical viewpoint the problem is why the localization length increases at the spin flop transition .
the first model for the localization length increase , invoking a three - dimensional ( 3d ) vrh mechanism , was proposed in ref . .
however , it is clear now that except for ultra - low temperatures ( that we estimate to be below @xmath15 ) , the vrh conduction at zero magnetic field is dominated by two - dimensional ( 2d ) physics .
@xcite because of this the 3d picture is not able to describe the most recent and detailed mr data , as we discuss below .
experiments are performed typically in the temperature range of a few kelvin and higher where the out - of - plane resistivity anisotropy is large @xmath16 . @xcite
while we ultimately expect that at @xmath17 vrh will become 3d , in the temperature range of experimental interest the 2d mechanism is the relevant one , as is clear from the analysis of the 2d-3d crossover temperature and the fits of the hopping conductivity presented in the next section . in the present work
we demonstrate that the large mr arises from a change of the effective dimensionality of the vrh mechanism with applied field .
we support our conclusions by detailed comparison with recent experiments on magnetotransport which can be described by our theory with excellent accuracy .
the main idea of the present work is that a dimensional crossover ( 2d @xmath3 3d ) occurs at the spin flop , and this is conceptually and quantitatively different from the 3d picture of ref . .
in particular in our approach the increase of the mr ( and the localization length ) is not simply due to the change of the out - of - plane effective mass as in ref . , but rather arises from a change in the shape of the ( localized ) wave - functions across the spin - flop .
in the temperature regime that we keep in mind , 1k and higher , the change of the out - of - plane effective mass is a small , secondary effect ( which can manifest itself only at ultra - low temperatures where the full 3d vrh mechanism is responsible for transport ) .
we show that the alignment of the weak ferromagnetic moments in neighboring planes with the field allows the inter - layer hopping of localized holes , which in turn leads to an increase of the hole s in - plane hopping probability and thus negative mr .
the presence of an inter - layer hopping channel across the spin - flop was already identified in ref . ; however our analysis differs in the effects this additional channel can produce in vrh conduction . by investigating the evolution of the hole
bound state as a function of magnetic field and temperature , we find that in various regimes different numbers of layers are involved in transport . in
the experimentally relevant temperature range the hopping turns out to be quasi - two - dimensional , leading to a negative mr in very good agreement with the most recent experiments .
@xcite the paper is organized as follows . in section
ii we analyze the effect of the magnetic field on the dispersion of the localized holes , through the inter - layer hopping . in section
iii we present a detailed analysis of the change of the hole s wave - function , due to the modified dispersion . in sections
iv and v we then use the wave - functions to calculate the magnetoresistance for out - of - plane and in - plane magnetic fields , and compare with experiment .
section vi contains our conclusions .
first we briefly summarize previous results related to the structure of the hole s 2d bound state at zero field .
as a starting point , we consider the hole dynamics in the antiferromagnetic background within the framework of the @xmath19 model .
@xcite in the absence of the coulomb potential @xmath20 of the sr ion , a hole resides , in momentum space , near the nodal points @xmath21 and has dispersion @xmath22 , where @xmath23 is the inverse 2d effective mass appropriate for lsco @xcite ( @xmath24 in absolute units ) .
we measure energies in units of @xmath25 , and the lattice spacing is set to unity . due to @xmath20
, the hole is localized , and its wave function has the form @xmath26 , corresponding to binding energy @xmath27 . here the inverse ( 2d ) localization radius for lsco is @xmath28 , @xcite giving @xmath29 . on a perfect square lattice
the bound state is four - fold degenerate : the hole can reside on either up or down sub - lattices ( pseudospin ) , and it can reside in either of the two pockets @xmath30 ( flavor ) .
the orthorhombic distortion lifts the flavor degeneracy due to the presence of diagonal next - nearest neighbor hopping @xmath31 .
hence , the holes occupy only the pocket @xmath32 . @xcite this is also consistent with the fact that the spin structure becomes incommensurate along the @xmath8 orthorhombic direction , as seen in neutron scattering for @xmath33 .
@xcite now let us consider the correction to the 2d dispersion @xmath34 arising from the inter - layer hopping @xmath35 . without account of correlations
we have @xmath36 , with @xmath37-dependent @xmath35 , @xmath38 , where @xmath39 .
@xcite by averaging over the momentum distribution ( @xmath40 ) in the 2d bound state , we find the effective value of inter - layer hopping @xmath41 .
since this is a crude , order of magnitude estimate , below we will use @xmath35 as a fitting parameter .
the @xmath42 model correlations change @xmath34 .
first , the hopping matrix element should be replaced by @xmath43 , where @xmath44 is the quasiparticle residue .
second , direct hopping is allowed only between spins in the same sublattice . in lsco a spin in a given plane ( e.g. spin 1 " in fig .
1 ) interacts with four others in the plane above ( and below ) , @xcite but at @xmath45 ( or @xmath46 ) only out - of - plane hopping in the @xmath47 direction is allowed , because this corresponds to ferromagnetic ordering of spins in neighboring planes ( see fig . 1 ) .
however , when a magnetic field is applied along the @xmath9-axis the spins in the next layer reverse their signs across the spin - flop transition at @xmath48 , so that for @xmath49 only hopping in the @xmath50 direction contributes .
this is schematically shown in fig . 1 , and reflects in the effective dispersion : @xmath51 where we define the z - direction @xmath52 .
since in the occupied pocket @xmath53 , @xmath54 , eq .
( [ zdispersion ] ) reads @xmath55 thus at @xmath56 there is no z - dispersion , the coherent dynamics is purely 2d and the vrh in - plane resistivity behaves as @xmath57 .
@xcite @xmath58 is strongly doping and sample dependent , and the data at zero field , @xmath59 can be fitted by @xmath60 and @xmath61 with astonishing accuracy in the range @xmath62k , as shown in fig .
[ fig2](inset ) . observe that a more accurate estimate of the z - dynamics below the spin flop can be obtained by expanding the in - plane dispersion @xmath63 around @xmath32 and minimizing the resulting quadratic form .
this gives a non - zero dispersion in the z - direction , but the corresponding effective mass is huge , @xmath64 \sim 10 ^ 4m_e$ ] .
consequently we find that at temperatures below @xmath65 , the vrh is ultimately three - dimensional , but this regime is irrelevant to present experiments . ; the blue arrows in the middle layer show the spin reversal for @xmath66 . ]
to qualitatively understand the effect of the change of dispersion ( [ dispersion2 ] ) at @xmath67 on the in - plane hole dynamics , we first estimate the change in the hole binding energy . from ( [ dispersion2 ] ) , after the spin flop the edge of the continuum ( @xmath68 ) decreases by @xmath69 , while the absolute energy of the bound state , to second order in the small parameter @xmath70 , decreases only by amount @xmath71 . the binding energy , which is the magnitude of the difference between the absolute energy and the continuum limit , changes after the flop as @xmath72 within the vrh picture the conduction is proportional to the hole s hopping probability , which decays exponentially away from the donor site due to the hole localization .
thus , the decrease of the hole binding energy ( [ binding ] ) across the spin flop signals an increase of the localization length , and in turn an increase of the vrh conductivity .
this is our central idea that explains the negative mr . to make this argument more quantitative , we need to compute the change in the hole s hopping probability across the spin - flop transition .
let us enumerate planes by the index @xmath73 and assume for simplicity that sr produces a _
local _ potential @xmath20 that acts only in the plane @xmath74 . for a shallow level
the exact form of the local potential is not important , @xcite and for simplicity we will use the @xmath75-function approximation : @xmath76 , where @xmath77 is assumed to be smaller than the localization length , @xmath78 .
the bound state is described by the wave function @xmath79 that depends on both @xmath73 and @xmath80 . before the spin flop , @xmath56
, it obeys the schrdinger equation : @xmath81 , where @xmath82 .
the solution for @xmath83 is given by @xmath84 , and for @xmath85 it reads @xmath86 where @xmath87 and @xmath88 are the modified bessel functions of the first and second kind respectively . for the inverse localization length
we obtain @xmath89 , where @xmath90 is euler s constant .
however the exact dependence of @xmath91 on the parameters of the potential is not important , since the energy scale that @xmath91 determines , the binding energy @xmath92 , was extracted from experiment in earlier work , @xmath29 ( see section ii ) .
for @xmath67 the schrdinger equation becomes @xmath93 = e\psi_n , \ ] ] with @xmath76 , and @xmath94 . after the fourier
transform @xmath95 we obtain @xmath96 by solving this equation we find @xmath97 thus , as we have already pointed out before eq .
( [ binding ] ) , after the spin flop the absolute energy @xmath98 is shifted only in the second order in @xmath99 and hence this shift can be neglected . consequently below
we set @xmath100 .
however , in contrast to eq .
( [ psi0n ] ) , the new wave function , @xmath101 does not have a simple exponential decay .
( solid , blue line ) across the spin - flop transition for @xmath102 .
here @xmath103 , @xmath61 .
the circles are data from ref . .
inset : zero field in - plane resistivity fit to 2d vrh form ( solid , red line ) with @xmath61 . ]
to evaluate the mr across the spin flop we compute now the change of the hole s probability to propagate at the ( large ) vrh distance @xmath104 : @xmath105 .
@xcite we can then identify two large - distance regimes for the wave function ( [ psin ] ) : ( 1 ) very large distances , @xmath106 , and ( 2 ) intermediate large distances , @xmath107 . in the first regime the integral in ( [ psin ] ) can be evaluated using the saddle - point approximation and the wave function is spread over many transverse channels , @xmath108 .
the probability density at distance @xmath80 , @xmath109 in this regime the wave function has a pure exponential decay with a localization length corresponding to the shift of the binding energy given in eq .
( [ binding ] ) .
we then find that the ratio of conductivities after ( @xmath67 ) and before ( @xmath56 ) the flop is @xmath110 here @xmath111 , with @xmath112 from eq.([psi0n ] ) .
this result is valid when @xmath113 , which happens in practice when the temperature is below @xmath114 . in this case
@xmath115 , i.e. the corresponding mr is large : @xmath116 , and a full 2d @xmath3 3d crossover is expected in the spin flop .
notice that the mr is always negative , @xmath117 . in the intermediate large distances regime , @xmath118 , the integral in ( [ psin ] ) can be evaluated by direct expansion in powers of @xmath119 .
only the layers @xmath120 contribute in this case , @xmath121 , leading to @xmath122 from ( [ prs ] ) we obtain ( @xmath123 ) across the spin flop @xmath124 this formula corresponds to a crossover from a pure 2d case to an intermediate dimension " ( three transverse channels ) , and it is justified when @xmath125 .
we have also performed a full numerical evaluation of @xmath126 and of the mr with @xmath79 from eq .
( [ psin ] ) , since the above considerations are based on asymptotic behavior .
the exact numerical form was used in both figures [ fig2 ] and [ fig3 ] below .
we have found quite clearly that indeed in the temperature range where experimental data are available , @xmath127 , the intermediate asymptotic formula ( [ slxs ] ) is the relevant one , since the mr is still relatively small there , e.g. @xmath128 .
however as the temperature is lowered , @xmath129 , the mr increases and the system enters a crossover region between the asymptotic formulas eq .
( [ slxs ] ) and eq .
( [ ult ] ) , which requires the use of the exact wave - functions .
the calculated mr is plotted in fig .
[ fig2 ] and we observe that the fitted value @xmath130 agrees quite well with the estimate @xmath131 from band structure calculations .
@xcite ( solid , blue line ) for in - plane field @xmath132 , with @xmath133 .
the squares are data from ref . .
inset : evolution of the angle ( [ angle ] ) with field . ]
in this case , due to the alignment of the dm - induced moments with the magnetic field , the spins rotate in the @xmath135 plane ( in opposite directions on the two sub - lattices ) , @xcite and align completely along @xmath9 at a field @xmath136 .
thus , once we introduce the angle @xmath137 that the spins make with the @xmath8-direction , our previous calculations leading to eq .
( [ slxs ] ) remain valid with the replacement @xmath138 .
observe that while for the undoped lco an intermediate in - plane spin flop is expected at @xmath139 , @xcite in lsco the doped holes contribute to enhance the @xmath8-axis spin susceptibility and then to confine the spins in the @xmath135 plane , as it has been discussed recently in ref . .
we can then write the field dependence of @xmath140 between @xmath45 and @xmath141 as @xmath142 in the above expression @xmath143 is the dm anisotropy , @xmath144 is the inter - layer exchange , @xmath145 is the out - of - plane ( or xy ) anisotropy gap , @xmath146 is the staggered order parameter , @xmath147 is the doping and @xmath148 is a dimensionless measure of the holes - induced spin susceptibility . here
@xmath149 is measured in units of @xmath150 , with @xmath151 and @xmath152 is the bohr magneton .
the parameter values can be extracted from the experiments , @xcite and for @xmath59 we take @xmath153 mev , @xmath154 ( mev)@xmath155 , @xmath156 mev , @xmath157 , and @xmath158 , which gives @xmath159 t. @xcite using @xmath160 as the only fitting parameter , we find a remarkable agreement with the experimental data of ref . at @xmath161 k , as shown in fig .
[ fig3 ] .
we also note that in oxygen - doped compounds the contribution of the localized holes to the longitudinal susceptibility is much smaller , with @xmath162 .
@xcite as a consequence , one expects to observe at a field @xmath163 t an intermediate flop which reflects in a kink in both the @xmath164 curve and in the mr curve , as it has been measured indeed in the earlier transport measurements in la@xmath165cuo@xmath166 .
in conclusion , we have shown that in the strongly localized vrh regime for @xmath167 , the in - plane mr is sensitive to the inter - layer hopping @xmath168 because an external magnetic field effectively changes the dimensionality of the problem , making the hopping quasi-2d .
the mr reflects the physics of the spin - flop and is always negative ; its value as well as temperature and field dependence are in excellent quantitative agreement with recent experiments in lsco at @xmath59 .
@xcite orbital effects , typically causing positive mr , @xcite are negligible at this small doping because the hole s localization length @xmath169 is much smaller than the magnetic length at any reasonable fields .
finally we comment that unlike lsco where the role of magnetic anisotropies is very well established , in insulating yba@xmath165cu@xmath170o@xmath171 ( ybco ) such anisotropies are expected to be very weak , which is related to the absence of strong localization in this material .
@xcite spin - related effects in the in - plane mr are absent for field in the @xmath9 direction , while the in - plane field mr remains small and appears to be due to the dynamics of holes that are very weakly influenced by disorder .
@xcite the complete understanding of these phenomena in ybco remains an open issue although it is clear that magnetotransport is not dominated by the local spin physics .
we are grateful to d. k. campbell , o. k. andersen , o. jepsen , a. lavrov , and y. ando for stimulating discussions .
o.p.s . gratefully acknowledges support from the alexander von humboldt foundation , and the hospitality of the max - planck - institute for solid state research stuttgart and leipzig university .
v.n.k . is supported by boston university .
is supported through nsf grant dmr-0343790 .
m. matsuda , m. fujita , k. yamada , r. j. birgeneau , m. a. kastner , h. hiraka , y. endoh , s. wakimoto , and g. shirane , phys .
b * 62 * , 9148 ( 2000 ) ; m. matsuda , m. fujita , k. yamada , r. j. birgeneau , y. endoh , and g. shirane , phys . rev .
b * 65 * , 134515 ( 2002 ) .
t. thio , t. r. thurston , n. w. preyer , p. j. picone , m. a. kastner , h. p. jenssen , d. r. gabbe , c. y. chen , r. j. birgeneau , and a. aharony , phys .
b * 38 * , 905 ( 1988 ) ; t. thio , c. y. chen , b. s. freer , d. r. gabbe , h. p. jenssen , m. a. kastner , p. j. picone , n. w. preyer , and r. j. birgeneau , phys . rev .
b * 41 * , 231 ( 1990 ) . l. benfatto and m. b. silva neto , phys . rev .
b * 74 * , 024415 ( 2006 ) ; l. benfatto , m. b. silva neto , a. gozar , b. s. dennis , g. blumberg , l. l. miller , s. komiya , and y. ando , phys . rev .
b * 74 * , 024416 ( 2006 ) . | we show that , due to the weak ferromagnetism of la@xmath0sr@xmath1cuo@xmath2 , an external magnetic field leads to a dimensional crossover 2d @xmath3 3d for the in - plane transport .
the crossover results in an increase of the hole s localization length and hence in a dramatic negative magnetoresistance in the variable range hopping regime .
this mechanism quantitatively explains puzzling experimental data on the negative magnetoresistance in the nel phase of la@xmath0sr@xmath1cuo@xmath2 . |
joy ( 1945 ) identified t tauri stars as a distinct class of variable stars and ambartsumian ( 1947 ; 1949 ) first suggested that they represent solar - like stars in the early stages of formation .
they have since that time become understood as such and have been studied vigorously ( e.g. menard & bertout 1999 and references therein ) .
the prototype and one of the brightest members of the class , t tau , was discovered in the mid 1800 s to vary significantly in visual magnitude . however , t tau is a complicated system and is not a typical member of the t tauri class .
three optical nebulosities are associated with it .
hind s variable nebula ( ngc 1555 ) and a small nebulosity ( ngc 1554 ) lie @xmath0 45@xmath1 and @xmath0 4@xmath2 to the west , respectively ( herbig 1950 ) .
the third , burnham s nebula , extends @xmath0 5@xmath1 from the star at a p.a .
= 152@xmath3 ( burnham 1890 ) .
multi - wavelength observations reveal complex distributions of gas and dust associated with the remnant protostellar envelope , molecular outflows ( van langevelde et al .
1994 ; momose et al . 1996 ; schuster et al .
1997 ) , arcs and filaments of material traced by molecular hydrogen ( herbst et al . 1996 ; herbst , robberto & beckwith 1997 ) and two herbig - haro jets ( bhm & solf 1994 ; solf & bhm 1999 ) .
t tau is a young triple system ; it consists of t tau , a k0 star at a v mag @xmath0 10 ( herbig & bell 1988 ) , and an infrared companion ( irc ) , 0.@xmath17 to the south ( dyck , simon & zuckerman 1982 ) , which is itself a binary with a projected separation of 0.@xmath105 ( koresko 2000 ) .
we monitored the near infrared flux of the system and found that the optically visible star does not vary significantly in magnitude in the k ( 2.2 @xmath4 m ) and l@xmath2 ( 3.8 @xmath4 m ) photometric bands ( beck et al .
2001 ; beck 2001 ) .
the irc binary does vary in near infrared flux on timescales as short as a week .
however , the irc has never been detected at wavelengths shortward of 1.2 @xmath4 m to a limiting v - band magnitude of 19.6 ( stapelfeldt et al .
1998 ) , and therefore does not contribute to the optical variability of the system .
lozinskii ( 1949 ) compiled the historical light curve of t tau from nearly 2000 observations spanning the interval 1858 to 1941 .
a translation of lozinskii s paper is available online at http://www.ess.sunysb.edu/2001aj/beck01.html . from 1858 to @xmath0 1915 , t tau dimmed and brightened randomly in visible light , between 10 and 14 mag in the visible .
after 1915 , t tau appeared mostly bright at @xmath0 10 mag .
lozinskii s light curve represents a smoothed 10-day average , and the sources of the individual measurements are unreferenced .
we were therefore interested to use the independent data of the photographic plates available in the harvard college observatory archives in order to determine the light curve of t tau to confirm its large light variations at the begining of the 20th century .
since the classical t tauri stars ry tau and rw aur appear on the same plates as t tau , we took advantage of the opportunity to measure their light curves as well .
we estimated the brightness of t tau , ry tau and rw aur by inspection of over 150 archival ac and am series photographic plates of the harvard college observatory archive .
these patrol plates were obtained using a telescope equipped with a 1.5-inch aperture cooke lens and provided data with a plate scale of 600@xmath1/mm spanning the time interval 1899 to 1952 .
we supplemented the t tau data with more than 125 measurements from the plates in the rh and rb series , which were obtained using 3-inch aperture ross lens at a scale of 390@xmath1/mm and covered the period 1928 to 1952 .
the spectral response of the photographic plates was essentially blue , so we estimated the magnitude of t tau , ry tau and rw aur by comparison with the b magnitudes of nearby stars .
the estimated magnitudes of our targets are therefore b magnitudes in this relative sense .
we first obtained the b - band magnitudes for three dozen stars within 1@xmath3 of t tau using the simbad database and usno catalog ( monet et al .
by identifiying these reference stars on plates spanning several decades we eliminated the obvious variables among them and determined the limiting sensitivities of the ac / am and rh / rb plates to be 12 - 13.5 mag and 14 - 15 mag , respectively .
we estimated the magnitude of t tau , ry tau and rw aur with reference to 16 stars with b magnitudes ranging from 9.6 to 13.8 in increments of 0.2 to 0.3 magnitudes .
the internal precision of the estimated magnitudes of the targets is @[email protected] mag and is determined by the magnitude spacing of the calibrators and the accuracy of their individual b - band magnitudes .
figures 1 through 3 present the derived light curves for t tau , ry tau and rw aur .
a table of the data used to generate these light curves is available online at the webpage referenced in @xmath6 1 .
inspection of digitized sky survey images revealed that the target stars are the brightest objects within a 3@xmath2 radius , hence we attribute the observed light fluctuations solely to them .
the light curve presented in figure 1 is similar to that determined by lozinskii ( 1949 ) in that the magnitude of t tau dimmed and brightened randomly in the early part of the 20th century from b @xmath0 11 to fainter than the detection limit on timescales as short as a month . to demonstrate the time scale of the observed light variations , figure 4
presents 15 measurements of the magnitude of t tau during 1902 . from 10 january to 10 february 1902 t tau brightened from 13 mag to @xmath011 mag , and six measurements in october - november 1902 show that it varied by @xmath01 mag on time scales of less than two weeks . a lomb normalized periodogram analysis ( press et al .
1994 ) of the light curve between 1899 and 1917 shows no significant periodicity on time scales of less than @xmath0 5 years . from the late 1910 s to 1952 , t tau stabilized at a b magnitude of @xmath0 11 and , aside from brief dimming events of 1.5 magnitudes in 1925 and 0.8 magnitudes in 1931 , did not vary further at a statistically significant level .
additional data on the variability of t tau , obtained from the online database of herbst et al .
( 1994 , hereafter h94 ) and the american association of variable star observers ( aavso ; mattei 2000 ) , shows that between 1937 and the present it has not varied by more than @xmath0 @xmath50.5 magnitudes from its average .
observations of the optical variability of t tau taken during a single week reveal that it presently does not vary by more than @xmath0 0.1 v - band magnitudes on this time scale ( ismailov 1997 ) .
the `` flickering '' of 2 - 3 magnitudes observed in t tau in the early years of 20th century was real and ceased abruptly in @xmath0 1917 .
the light curve of ry tau ( figure 2 ) is characterized by two time scales .
it varies by 2 - 3 magnitudes over approximately a decade and by @xmath0 1 magnitude on time scales of less than a year .
this type of variability has continued to the present ( h94 ; holtzman , herbst & booth 1986 ; petrov et al .
rw aur varied by 2 - 3 magnitudes on timescales as short as a month throughout the 1899 - 1952 interval ( figure 3 ) .
report similar variability ; apparently this behavior has continued for at least a century .
although the amplitudes and time scales of the light fluctuations of rw aur and t tau before 1917 are similar , we argue in @xmath64 that their variability is probably caused by different mechanisms .
parenago ( 1954 ) described a classification which is useful to categorize the light curves of t tauri stars ( see also herbig 1962 ) : based on this scheme , t tau was a class iv variable from 1899 to 1917 , and a class i from 1917 to the present .
ry tau and rw aur are class iv variables during the interval we have sampled , although breaking the ry tau data into sections , as in figure 2 , suggests that its variability may be migrating from one class to another on decade - long time scales . parenago s classification is useful because it draws attention to the fact that the character of t tau s variability changed significantly and abruptly .
it does not , however , provide interpretation of the causes of the variability .
in their study of the light and color variability of @xmath0 80 t tauri stars , h94 argued that the variability of most t tauris is produced by one , or a combination , of three distinct mechanisms : photometric and spectroscopic monitoring studies show that the optical variability of ry tau is likely caused by variable obscuration ( holtzman , herbst & booth 1986 ; petrov et al .
1998 ) , thus it is best described as a type iii variable .
h94 argued that t tau is the only star in their sample that shows characteristics of all three variability mechanisms .
herbst et al . ( 1986 ) discovered rotational modulation of t tau s light with a period of 2.8 days .
the amplitude of the rotational modulation is small , @xmath0 0.02 mag , so it seems very unlikely that this modulatation could account for the 2 - 3 magnitude fluctuations in the early 1900 s .
the results of herbst et al .
( 1982 ) and basri & bathala ( 1990 ) show that the light fluctuations of rw aur correlate with h@xmath7 and veiling emission .
the variability of rw aur is therefore best classified as type ii ( h94 ) .
although the amplitude and timescale of the light variations of t tau in the early years of the 20th century are similar to those of rw aur , it is improbable that variable accretion was responsible for t tau s variability .
it is very difficult to imagine how the accretion rate , after varying significantly enough to produce light fluctuations of 2 - 3 magnitudes , could stabilize in a high state with a sufficiently constant amplitude to produce light variations of less 0.5 magnitudes around b @xmath0 11 .
the most likely explanation for the observed variability of t tau prior to 1917 is variable extinction along the line of sight to the star .
t tau is now obscured by a@xmath8 @xmath0 1 ( cohen & kuhi 1979 ) , but clumpy material with 2 - 3 magnitudes of extinction moving through the line of sight to the star could explain the observed dimmings
. the eventual clearing of this material would result in a more or less constant light output .
hogerheijde et al ( 1997 ) and akeson et al .
( 1998 ) detect a circumstellar disk around t tau by its emission at mm wavelengths and conclude that it is observed nearly face - on .
variations in extinction toward t tau are therefore not likely to be caused by material in its own disk , unless it is strongly warped .
weintraub et al . ( 1989 ) found evidence for keplerian rotation in material surrounding t tau , and momose et al .
( 1996 ) , schuster et al . (
1997 ) , and weintraub et al .
( 1999 ) detected complex distributions of material associated with the t tau system .
two herbig - haro flows are associated with t tau ; hh 155 originates from the optically visible star , and hh 355 from the irc ( bhm & solf 1994 ; herbst , robberto & beckwith 1997 ; solf & bhm 1999 ) .
hence , it is conceivable that material associated with the envelope or outflows in the t tau system could have caused the observed extinction variability of t tau .
the `` flickering '' and abrupt clearing of the obscuring material implies significant density inhomogeneities on spatial scales of a few stellar radii .
it is reasonable to speculate that the possible variation in extinction presently observed toward the t tau irc ( beck , prato & simon 2001 ) is caused by a similar distribution of material responsible for the optical variability of t tau in the early 1900 s . in a region as complex as that of t tau
, it seems impossible to identify now the material that would have produced @xmath03 magnitudes of extinction a century ago . a reasonable scenario for the variable obscuration must also explain the abrupt cessation and apparent non - repeatability of the variability .
it is worth noting that the components of the t tau triple system were likely positioned differently a century ago .
roddier et al .
( 2000 ) and ghez et al .
( 1995 ) monitored the orbital motion of the t tau north - south system from 1989 to 1999 and found that the position angle of t tau s has changed by @xmath06@xmath3 with respect to t tau n , but that it has not varied significantly in angular separation ( @xmath9 0.@xmath103 ) .
hence , the period of the t tau n - s system , at a 100 au separation , is likely on the order of @xmath0500 - 700 years .
the variability may have been the result of the different geometry of obscuring material associated with the disks or outflows of the system components . like
t tau s historic variability , we speculate that other puzzling aspects of t tauri behavior , such as the discovery of the irc to uy aur when it was bright in the visible ( joy & van biesbroeck 1944 ) , may have extrinsic rather than intrinsic causes . a more complete understanding of the orbital motion and orientation of the disks and outflows of these systems will enable tests of this idea .
we thank g. herbig for conversations that led to this study and for valuable advice in its course .
we also thank m. hazen and a. doane for their assistance and hostpitality during our use of the harvard college observatory plate archives , w. herbst for providing the database of t tauri star variability , w. sherry for his assistance locating potential calibration stars , s. wolk for his advice regarding periodogram analysis techniques , d. beznosko for help with the translation of lozinskii s article , and the referee for a helpful report .
we have used , and acknowledge with thanks , data from the aavso international database , based on observations submitted to the aavso by variable star observers worldwide .
this work also made use of the simbad database , operated at cds .
strasbourg , france , and the digitized sky surveys which were produced at the space telescope science institute under u.s .
government grant nag w-2166 .
our work was supported in part by nsf grant 98 - 19694 .
beck , t. l. , simon , m. , ghez , a. m. , prato , l. & howell , r. r. 2000 _ the near infrared & ice - b & variability of t tau & haro 6 - 10 _ in _ the birth & evolution of binary stars _ poster proceedings of the iau symposium no .
200 , eds . b.
reipurth & h. zinnecker , 51 roddier , f. , roddier , c. , brandner , w. , charissoux , d. , veran , j .-
courbin , f. 2000 , _ five - years of adaptive optics observations of t tau south _ in _ the birth & evolution of binary stars _
poster proceedings of the iau symposium no .
200 , eds .
bo reipurth & hans zinnecker , 60 | we present the historical light curve of t tau derived from photographic plates in the harvard college observatory archives .
we find that the optical light of t tau varied by 2 - 3 ( or more ) magnitudes on time scales as short as a month prior to @xmath0 1917 , consistent with the results of lozinskii ( 1949 ) .
extreme light fluctuations of greater than 2 magnitudes abruptly ceased in the late 1910 s and , to the best of our knowledge , have not repeated since this time .
we compare the observed light variations of t tau to the t tauri stars ry tau and rw aur , whose light curves we also constructed from inspection of the archival plates .
we find that variable extinction along the line of sight to the star is the most likely explanation for the observed light fluctuation of t tau during the early part of the 20th century . |
-5 mm how can we understand intelligent behavior ?
how can we design intelligent computers ?
these are questions that have been discussed by scientists and the public at large for over 50 years . as mathematicians ,
however , the question we want to ask is `` is there a _ mathematical _ theory underlying intelligence ? ''
i believe the first mathematical attack on these issues was control theory , led by wiener and pontryagin .
they were studying how to design a controller which drives a motor affecting the world and also sits in a feedback loop receiving measurements from the world about the effect of the motor action .
the goal was to control the motor so that the world , as measured , did something specific , i.e. move the tiller so that the boat stays on course .
the main complication is that nothing is precisely predictable : the motor control is not exact , the world does unexpected things because of its complexities and the measurements you take of it are imprecise .
all this led , in the simplest case , to a beautiful analysis known as the wiener - kalman - bucy filter ( to be described below ) .
but control theory is basically a theory of the output side of intelligence with the measurements modeled in the simplest possible way : e.g. linear functions of the state of the world system being controlled plus additive noise .
the real input side of intelligence is perception in a much broader sense , the analysis of all the noisy incomplete signals which you can pick up from the world through natural or artificial senses .
such signals typically display a mix of distinctive patterns which tend to repeat with many kinds of variations and which are confused by noisy distortions and extraneous clutter .
the interesting and important structure of the world is thus coded in these signals , using a code which is complex but not perversely so .
-5 mm the first serious attack on problems of perception was the attempt to recognize speech which was launched by the us defense agency arpa in 1970 .
at this point , there were two competing ideas of what was the right formalism for combining the various clues and features which the raw speech yielded .
the first was to use logic or , more precisely , a set of ` production rules ' to augment a growing database of true propositions about the situation at hand .
this was often organized in a ` blackboard ' , a two - dimensional buffer with the time of the asserted proposition plotted along the @xmath0-axis and the level of abstraction ( i.e. signal phone phoneme syllable word sentence ) along the @xmath1-axis .
the second was to use statistics , that is , to compute probabilities and conditional probabilities of various possible events ( like the identity of the phoneme being pronounced at some instant ) .
these statistics were computed by what was called the ` forward - backward ' algorithm , making 2 passes in time , before the final verdict about the most probable translation of the speech into words was found .
this issue of logic vs. statistics in the modeling of thought has a long history going back to aristotle about which i have written in [ m ] .
i think it is fair to say that statistics won .
people in speech were convinced in the 1970 s , artificial intelligence researchers converted during the 1980 s as expert systems needed statistics so clearly ( see pearl s influential book [ p ] ) , but vision researchers were not converted until the 1990 s when computers became powerful enough to handle the much larger datasets and algorithms needed for dealing with 2d images .
the biggest reason why it is hard to accept that statistics underlies all our mental processes perception , thinking and acting is that we are not consciously aware of 99% of the ambiguities with which we deal every second .
what philosophers call the ` raw qualia ' , the actual sensations received , do not make it to consciousness ; what we are conscious of is a precise unambiguous enhancement of the sensory signal in which our expectations and our memories have been drawn upon to label and complete each element of the percept .
a very good example of this comes from the psychophysical experiments of warren & warren [ w ] in 1970 : they modified recorded speech by replacing a single phoneme in a sentence by a noise and played this to subjects . remarkably ,
the subjects did _ not _ perceive that a phoneme was missing but believed they had heard the one phoneme which made the sentence semantically consistent : l|l actual sound & perceived words + the ?
eel is on the shoe & the _ h_eel is on the shoe + the ?
eel is on the car & the _ wh_eel is on the car + the ?
eel is on the table & the _ m_eal is on the table + the ?
eel is on the orange & the _ p_eel is on the orange two things should be noted .
firstly , this showed clearly that the actual auditory signal did not reach consciousness .
secondly , the choice of percept was a matter of probability , not certainty .
that is , one might find some odd shoe with a wheel on it , a car with a meal on it , a table with a peel on it , etc .
but the words which popped into consciousness were the most likely .
an example from vision of a simple image , whose contents require major statistical reasoning to reconstruct , is shown in figure [ fig : oldman ] .
it is important to clarify the role of probability in this approach .
the uncertainty in a given situation need not be caused by observations of the world being truly unpredictable as in quantum mechanics or even effectively so as in chaotic phenomena .
it is rather a matter of efficiency : in order to understand a sentence being spoken , we do not need to know all the things which affect the sound such as the exact acoustics of the room in which we are listening , nor are we even able to know other factors like the state of mind of the person we are listening to . in other words , we always have incomplete data about a situation .
a vast number of physical and mental processes are going on around us , some germane to the meaning of the signal , some extraneous and just cluttering up the environment . in this ` blooming , buzzing ' world , as william james called it , we need to extract information and the best way to do it , apparently , is to make a stochastic model in which all the irrelevent events are given a simplified probability distribution .
this is not unlike the stochastic approach to navier - stokes , where one seeks to replace turbulence or random molecular effects on small scales by stochastic perturbations .
-5 mm having accepted that we need to use probabilities to combine bits and pieces of evidence , what is the mathematical set up for this ?
we need the following ingredients : a ) a set of random variables , some of which describe the observed signal and some the ` hidden ' variables describing the events and objects in the world which are causing this signal , b ) a class of stochastic models which allow one to express the variability of the world and the noise present in the signals and c ) specific parameters for the one stochastic model in this class which best describes the class of signals we are trying to decode now .
more formally , we shall assume we have a set @xmath2 of observed and hidden random variables , which may have real values or discrete values in some finite or countable sets , we have a set @xmath3 of parameters and we have a class of probability models @xmath4 on the @xmath0 s for each set of values of the @xmath5 s .
the crafting or learning of this model may be called the first problem in the mathematical theory of perception .
it is usual to factor these probability distributions : @xmath6 where the first factor , describing the likelihood of the observations from the hidden variables , is called the _ imaging model _ and the second , giving probabilities on the hidden variables , is called the _
prior_. in the full bayesian setting , one has an even stronger prior , a full probability model @xmath7 , including the parameters .
the second problem of perception is that we need to estimate the values of the parameters @xmath5 which give the best stochastic model of this aspect of the world .
this often means that you have some set of measurements @xmath8 and seek the value of @xmath5 which maximizes their likelihood @xmath9 .
if the hidden variables as well as the observations are known , this is called supervised learning ; if the hidden variables are not known , then it is unsupervised and one may maximize , for instance , @xmath10 .
if one has a prior on the @xmath5 s too , one can also estimate them from the mean or mode of the full posterior @xmath11 .
usually a more challenging problem is how many parameters @xmath5 to include . at one extreme
, there are simple ` off - the - shelf ' models with very few parameters and , at the other extreme , there are fully non - parametric models with infinitely many parameters . here
the central issue is how much data one has : for any set of data , models with too few parameters distort the information the data contains and models with too many overfit the accidents of this data set .
this is called the _ bias - variance dilemma_. there are two main approaches to this issue .
one is cross - validation : hold back parts of the data , train the model to have maximal likelihood on the training set and test it by checking the likelihood of the held out data .
there is also a beautiful theoretical analysis of the problem due principally to vapnik [ v ] and involving the _ vc dimension _ of the models the size of the largest set of data which can be split in all possible ways into more and less likely parts by different choices of @xmath5 .
as grenander has emphasized , a very useful test for a class of models is to synthesize from it , i.e. choose random samples according to this probability measure and to see how well they resemble the signals we are accustomed to observing in the world .
this is a stringent test as signals in the world usually express layers and layers of structure and the model tries to describe only a few of these . the third problem of perception is using this machinary to actually perceive : we assume we have measured specific values @xmath12 and want to infer the values of the hidden variables @xmath13 in this situation . given these observations , by bayes rule ,
the hidden variables are distributed by the so - called _ posterior _ distribution : @xmath14 one may then want to estimate the mode of the posterior , the most likely value of @xmath13 . or one may want to estimate the mean of some functions @xmath15 of the hidden variables . or , if the posterior is often multi - modal and some evidence is expected to available later , one usually wants a more complete description or approximation to the full posterior distribution .
-5 mm a convenient way to introduce the ideas of pattern theory is to outline the simple hidden markov model method in speech recognition to illustrate many of the ideas and problems which occur almost everywhere . here
the observed random variables are the values of the sound signal @xmath16 , a pressure wave in air .
the hidden random variables are the states of the speaker s mouth and throat and the identity of the phonemes being spoken at each instant .
usually this is simplified , replacing the signal by samples @xmath17 and taking for hidden variables a sequence @xmath18 whose values indicate which phone in which phoneme is being pronounced at time @xmath19 .
the stochastic model used is : @xmath20 i.e. the @xmath21 form a markov chain and each @xmath22 depends only on @xmath18 .
this is expressed by the graph : @xmath23 in which each variable corresponds to a vertex and the graphical markov property holds : if 2 vertices @xmath24 in the graph are separated by a subset @xmath25 of vertices , then the variables associated to @xmath26 and @xmath27 are conditionally independent if we fix the variables associated to @xmath25 . this simple model works moderately well to decode speech because of the linear nature of the graph , which allows the ideas of dynamic programming to be used to solve for the marginal distributions and the modes of the hidden variables , given any observations @xmath28 .
this is expressed simply in the recursive formulas : @xmath29 note that if each @xmath18 can take @xmath30 values , the complexity of each time step is @xmath31 . in any model ,
if you can calculate the conditional probabilities of the hidden variables and if the model is of exponential type , i.e.@xmath32 then there is also an efficient method of optimizing the parameters @xmath5 .
this is called the _ em algorithm _ and , because it holds for hmm s , it is one of the key reasons for the early successes of the stochastic approach to speech recognition .
for instance , a markov chain @xmath21 is an exponential model if we let the @xmath5 s be @xmath33 and write the chain probabilities as : @xmath34 the fundamental result on exponential models is that the @xmath5 s are determined by the expectations @xmath35 and that any set of expectations @xmath36 that can be achieved in some probability model ( with all probabilities non - zero ) , is also achieved in an exponential model .
-5 mm in this model , the observations @xmath22 are naturally continuous random variables , like all primary measurements of the physical world . but the hidden variables are discrete : the set of phonemes , although somewhat variable from language to language , is always a small discrete set .
this combination of discrete and continuous is characteristic of perception .
it is certainly a psychophysical reality : for example experiments show that our perceptions lock onto one or another phoneme , resisting ambiguity ( see [ l ] , ch.8 , esp .
but it shows itself more objectively in the low - level statistics of natural signals .
take almost any class of continuous real - valued signals @xmath16 generated by the world and compile a histogram of their changes @xmath37 over some fixed time interval @xmath38 .
this empirical distribution will very likely have kurtosis ( @xmath39 ) greater than 3 , the kurtosis of any gaussian distribution !
this means that , compared to a gaussian distribution with the same mean and standard deviation , @xmath0 has higher probability of being quite small or quite large but a lower probability of being average .
thus , compared to brownian motion , @xmath16 tends to move relatively little most of the time but to make quite large moves sometimes .
this can be made precise by the theory of stochastic processes with iid increments , a natural first approximation to any stationary markov process .
the theory of such processes says that ( a ) their increments always have kurtosis at least 3 , ( b ) if it equals 3 the process is brownian and ( c ) if it is greater , samples from the process almost surely have discontinuities . at the risk of over - simplfying
, we can say _
kurtosis @xmath40 3 is nature s universal signal of the presence of discrete events / objects in continuous space - time_. a classic example of this are stock market prices .
their changes ( or better , changes in log(price ) ) have a highly non - gaussian distribution with polynomial tails . in speech , the changes in the log(power ) of the windowed fourier transform show the same phenomenon , confirming that @xmath16 can not be decently modeled by colored gaussian noise .
-5 mm applying hmm s in realistic settings , it usually happens that @xmath30 is too large for an exhaustive search of complexity @xmath31 or that the @xmath18 are real valued and , when adequately sampled , again @xmath30 is too large .
there is one other situation in which the hmm - style approach works easily the _ kalman filter_. in kalman s setting , each variable @xmath18 and @xmath22 is real vector - valued instead of being discrete and @xmath41 and @xmath42 are _ gaussian distributions with fixed covariances and means depending linearly on the conditioning variable_. it is then easy to derive recursive update formulas , similar to those above , for the conditional distributions on each @xmath18 , given the past data @xmath43 . but usually , in the real - valued variable setting , the @xmath44 s are more complex than gaussian distributions .
an example is the tracking problem in vision : the position and velocity @xmath18 of some specific moving object at time @xmath45 is to be inferred from a movie @xmath46 , in which the object s location is confused by clutter and noise .
it is clear that the search for the optimal reconstruction @xmath18 must be pruned or approximated .
a dramatic breakthrough in this and other complex situations has been to adapt the hmm / kalman ideas by using weak approximations to the marginals @xmath47 by a finite set of samples , an idea called _ particle filtering _ :
@xmath48 this idea was proposed originally by gordon , salmond and smith [ g - s - s ] and is developed at length in the recent survey [ d - f - g ] .
an example with explicit estimates of the posterior from the work of isard and blake [ i - b ] is shown in figure 2 .
they follow the version known as bootstrap particle filtering in which , for each @xmath49 , @xmath30 samples @xmath50 are drawn with replacement from the weak approximation above , each sample is propagated randomly to a new sample @xmath51 at time @xmath52 using the prior @xmath53 and these are reweighted proportional to @xmath54 .
-5 mm a more serious problem with the hmm approach is that the markov assumption is never really valid and it may be much too crude an approximation . consider speech recognition .
the finite lexicon of words clearly constrains the expected phoneme sequences , i.e. if @xmath18 are the phonemes , then @xmath55 depends on the current word(s ) containing these phonemes , i.e. on a short but variable part of the preceding string @xmath56 of phonemes . to fix this
, we could let @xmath18 be a pair consisting of a word and a specific phoneme in this word ; then @xmath55 would have two quite different values depending on whether @xmath57 was the last phoneme in the word or not . within a word
, the chain needs only to take into account the variability with which the word can be pronounced . at word boundaries
, it should use the conditional probabilities of word pairs .
this builds much more of the patterns of the language into the model .
why stop here ?
state - of - the - art speech recognizers go further and let @xmath18 be a pair of consecutive words plus a triphone in the second word ( or bridging the first and second word ) whose middle phoneme is being pronounced at time @xmath45 .
then the transition probabilities in the hmm involve the statistics of ` trigrams ' , consecutive word triples in the language .
but grammar tells us that words sequences are also structured into phrases and clauses of variable length forming a parse tree .
these clearly affect the statistics .
semantics tells us that words sequences are further constrained by semantic plausibility ( ` sky ' is more probable as the word following ` blue ' than ` cry ' ) and pragmatics tells us that sentences are part of human communications which further constrain probable word sequences .
all these effects make it clear that certain parts of the signal should be grouped together into units on a higher level and given labels which determine how likely they are to follow each other or combine in any way .
this is the essence of grammar : higher order random variables are needed whose values are subsets of the low order random variables .
the simplest class of stochastic models which incorporate variable length random substrings of the phoneme sequence are _ probabilistic context free grammars _ or pcfg s .
mathematically , they are a particular type of random branching tree . * definition * _ a is a stochastic model in which the random variables are ( a ) a sequence of rooted trees @xmath58 , ( b ) a linearly ordered sequence of observations @xmath22 and a 1:1 correspondence between the observations @xmath22 and the leaves of the whole forest of trees such that the children of any vertex of any tree form an interval @xmath59 in time and ( c ) a set of labels @xmath60 for each vertex .
the probability model is given by conditional probabilities @xmath61 for the labels of each child of each vertex is the name of the ` production rule ' with this vertex as its head , esp .
it fixes the arity of the vertex .
we are doing it this way to simplify the markov property . ] and @xmath62 for the observations , conditional on the label of the corresponding leaf .
_ see figure 3 for an example .
this has a markov property if we define the ` extended ' state @xmath63 at leaf @xmath49 to be not only the label @xmath18 at this leaf but the whole sequence of labels on the path from this leaf to the root of the tree in which this leaf lies .
conditional on this state , the past and the future are independent .
this is a mathematically elegant and satisfying theory : unfortunately , it also fails , or rather explodes because , in carrying it out , the set of labels gets bigger and bigger .
for instance , it is not enough to have a label for noun phrase which expands into an adjective plus a noun .
the adjective and noun must agree in number and ( in many languages ) gender , a constraint that must be carried from the adjective to the noun ( which need not be adjacent ) via the label of the parent .
so we need 4 labels , all combinations of singular / plural masculine / feminine noun phrases .
and semantic constraints , such as pr(`blue sky ' ) @xmath40 pr(`blue cry ' ) , would seem to require even more labels like ` colorable noun phrases ' .
rather than letting the label set explode , it is better to consider a bigger class of grammars , which express these relations more succinctly but which are not so easily converted into hmm s : _ unification grammars _ [ sh ] or _ compositional grammars _ [ b - g - p ] .
the need for grammars of this type is especially clear when we look at formalisms for expressing the grouping laws in vision : see figure 3 .
the further development of stochastic compositional grammars , both in language and vision , is one of the main challenges today .
-5 mm the theory of hmm s deals with one - dimensional signals .
but images , the signals occurring in vision , are usually two - dimensional or three - dimensional for mr scans and movies ( 3 space dimensions and 2 space plus 1 time dimension ) , even four - dimensional for echo cardiograms . on the other hand ,
the parse tree is a more abstract graphical structure and other ` signals ' , like medical data gathered about a particular patient , are structured in complex ways ( e.g. a set of blood tests , a medical history ) .
this leads to the basic insight of grenander s pattern theory [ g ] : that the variables describing the structures in the world are typically related in a graphical fashion , edges connecting variables which have direct bearing on each other .
finding the right graph or class of graphs is a crucial step in setting up a satisfactory model for any type of patterns .
thus the applications , as well as the mathematical desire to find the most general setting for this theory , lead to the idea of replacing a simple chain of variables by a set of variables with a more general graphical structure .
the general concept we need is that of a markov random field : * definition * _ a is a graph @xmath64 , a set of random variables @xmath65 , one for each vertex , and a joint probability distribution on these variables of the form : @xmath66 where @xmath67 ranges over the cliques ( fully connected subsets ) of the graph , @xmath68 are any functions and @xmath69 a constant . if the variables @xmath60 are real - valued for @xmath70 , we make this into a probability density , multiplying by @xmath71 .
moreover , we can put each model in a family by introducing a temperature @xmath72 and defining : _
@xmath73 these are also called _
gibbs models _ in statistical mechanics ( where the @xmath68 are called _ energies _ ) and _ graphical models _ in learning theory and , like markov chains , are characterized by their conditional independence properties .
this characterization , called the hammersley - clifford theorem , is that if two vertices @xmath74 are separated by a subset @xmath75 ( all paths in @xmath76 from @xmath26 to @xmath27 must include some vertex in @xmath25 ) , then @xmath77 and @xmath78 are conditionally independent given @xmath79 .
the equivalence of these independence properties , plus the requirement that all probabilities be positive , with the simple explicit formula for the joint probabilities makes it very convincing that mrf s are a natural class of stochastic models .
-5 mm this class of models is very expressive and many types of patterns which occur in the signals of nature can be captured by this sort of stochastic model . a basic example is the ising model and its application to the image segmentation problem . in the simplest form ,
we take the graph @xmath76 to be a square @xmath80 grid with two layers , with observable random variables @xmath81 associated to the top layer and hidden random variables @xmath82 associated to the bottom layer .
we connect by edges each @xmath83 vertex to the @xmath84 vertex above it and to its 4 neighbors @xmath85 in the @xmath0-grid ( except when the neighbor is off the grid ) and no others .
the cliques are just the pairs of vertices connected by edges . finally , we take for energies : @xmath86 the modes of the posteriors @xmath87 are quite subtle : @xmath0 s at adjacent vertices try to be equal but they also seek to have the same sign as the correponding @xmath88 .
if @xmath88 has rapid positive and negative swings , these are in conflict .
hence the more probable values of @xmath0 will align with the larger areas where @xmath88 is consistently of one sign .
this can be used to model a basic problem in vision : the _ segmentation problem_. the vision problem is to decompose the domain of an image @xmath1 into parts where distinct objects are seen .
for example , the oldman image might be decomposed into 6 parts : his body , his head , his cap , the bench , the wall behind him and the sky .
the decomposition is to be based on the idea that the image will tend to either slowly varying or to be statistically stationary at points on one object , but to change abruptly at the edges of objects . as proposed in [ g - g ] , the ising model can be used to treat the case where the image has 2 parts , one lighter and one darker , so that at the mode of the posterior the hidden variables @xmath0 will be @xmath89 on one part , @xmath90 on the other .
an example is shown in figure 4 .
this approach makes a beautiful link between statistical mechanics and perception , in which the process of finding global patterns in a signal is like forming large scale structures in a physical material as the temperature cools through a phase transition .
+ more complex models of this sort have been used extensively in image analysis , for texture segmentation , for finding disparity in stereo vision , for finding optic flow in moving images and for finding other kinds of groupings .
we want to give one example of the expressivity of these models which is quite instructive .
we saw above that exponential models can be crafted to reproduce some set of observed expectations but we also saw that scalar statistics from natural signals typically have high kurtosis , i.e. significant outliers , so that their whole distribution and not just their mean needs to be captured in the model .
putting these 2 facts together suggests that we seek exponential models which duplicate the whole distribution of some important statistics @xmath91 .
this can be done using as parameters not just unknown constants but unknown functions : @xmath92 if @xmath93 depends only the variables @xmath94 , for some clique @xmath95 , this is a mrf , whose energies have unknown functions in them .
an example of this fitting is shown in figure 5 .
-5 mm however , a problem with mrf models is that the dynamic programming style algorithm used in speech and one - dimensional models to find the posterior mode has no analog in 2d .
one strategy for dealing with this , which goes back to metropolis , is to imitate physics and introduce an artifical dynamics into the state space whose equilibrium is the gibbs distribution .
this dynamics is called a _
monte carlo markov chain _
( mcmc ) and is how the panels in figure 4 were generated .
letting the temperature converge to zero , we get _ simulated annealing _ ( see [ g - g ] ) and , if we do it slowly enough , will find the mode of the mrf model .
although slow , this can be speeded up by biasing the dynamics ( called _ importance sampling _
see [ t - z ] for a state - of - the - art implementation with many improvements ) and is an important tool .
recently , however , another idea due to weiss and collaborators ( see [ y - f - w ] ) and linked to statistical mechanics has been found to give new and remarkably effective algorithms for finding these modes . from an algorithmic standpoint ,
the idea is to use the natural generalization of dynamic programming , called _ bayesian belief propagation _ ( bbp ) , which computes the marginals and modes correctly whenever the graph is a tree and just use it anyway on an arbitrary graph @xmath76 !
mathematically , it amounts to working on the universal covering graph @xmath96 , which is a tree , hence much simpler , instead of @xmath76 . in statistical mechanics
, this idea is called the _ bethe approximation _ , introduced by him in the 30 s .
to explain the idea , start with the _ mean field approximation_. the mean field idea is to find the best approximation of the mrf @xmath44 by a probability distribution in which the variables @xmath60 are all independent .
this is formulated as the distribution @xmath97 which minimizes the kullback - liebler divergence @xmath98 . unlike computing the true marginals of @xmath44 on each @xmath60 which is very hard ,
this approximation can be found by solving iteratively a coupled set of non - linear equations for the @xmath99 .
but the assumption of independence is much too restrictive .
the idea of bethe is instead to approximate @xmath44 by a @xmath100-invariant distribution on @xmath96 .
such distributions are easy to describe : note that a markov random field on a _ tree _ is uniquely determined by its marginals @xmath101 for each edge @xmath102 and , conversely , if we are given a compatible set of distributions @xmath103 for each edge ( in the sense that , for all edges @xmath104 abutting a vertex @xmath105 , the marginals of @xmath106 give distributions on @xmath105 independent of @xmath49 ) , they define an mrf on @xmath76 .
so if we start with a markov random field on any @xmath76 , we get a @xmath100-invariant markov random field on @xmath96 by making duplicate copies for each random variable @xmath107 for each @xmath108 over @xmath105 and lifting the edge marginals .
but more generally , if we have any compatible set of probability distributions @xmath109 on @xmath76 , we also get a @xmath100-invariant mrf on @xmath96 .
then the bethe approximation is that family @xmath110 which minimizes @xmath111 . as in the mean field case
, there is a natural iterative method of solving for this minimum , which turns out , remarkably , to be identical to the generalization of bbp to general graphs @xmath76 .
this approach has proved effective in some cases at finding best segmentations of images via the mode of a two - dimensional mrf .
other interesting ideas have been proposed for solving the segmentation problem which we do not have time to sketch : region growing , see esp .
[ z - y ] ) , using the eigenfunctions of the graph - theoretic laplacian , see [ s - m ] , and multi - scale algorithms , see [ p - b ] and [ s - b - b ] .
-5 mm although signals as we measure them are always sampled discretely , in the world itself signals are functions on the continua , time or space or both together . in some situations , a much richer mathematical theory emerges by replacing a countable collection of random variables by random processes and asking whether we can find good stochastic models for these continuous signals .
i want to conclude this talk by mentioning three instances where some interesting analysis has arisen when passing to the continuum limit and going into some detail on two . we will not worry about algorithmic issues for these models .
-5 mm this is the area where the most work has been done , both because of its links with other areas of analysis and because it is one of the central problems of image processing . you observe a degraded image @xmath112 as a function of continuous variables and seek to restore it , removing simultaneously noise and blur . in the discrete
setting , the ising model or variants thereof discussed above can be applied for this .
there are two closely related ways to pass to the continuous limit and reformulate this as a problem in analysis .
as both drop the stochastic interpretation and have excellent treatments in the literature , we only mention briefly one of a family of variants of each approach : _ optimal piecewise smooth approximation of i via a variational problem _ : @xmath113 where @xmath114 , the improved image , has discontinuities along the set of ` edge ' curves @xmath115 .
this approach is due to the author and shah and has been extensively pursued by the schools of degiorgi and morel .
see [ m - s ] .
it is remarkable that it is still unknown whether the minima to this functional are well behaved , e.g. whether @xmath115 has a finite number of components .
stochastic variants of this approach should exist .
_ non - linear diffusion of i _ : @xmath116 where @xmath114 at some future time is the enhancement .
this approach started with the work of perona and malik and has been extensively pursued by osher and his coworkers .
see [ gu - m ] .
it can be interpreted as gradient descent for a variant of the previous variational problem .
-5 mm one of the earliest discoveries about the statistics of images @xmath117 was that their power spectra tend to obey power laws @xmath118 where @xmath119 varies somewhat from image to image but clusters around the value 2 .
this has a very provocative interpretation : this power law is implied by self - similarity ! in the language of lattice field theory , if @xmath120 is a random lattice field and @xmath121 is the block averaged field @xmath122 then we say the field is a renormalization fixed point if the distributions of @xmath117 and of @xmath121 are the same .
the hypothesis that natural images of the world , treated as a single large database , have renormalization invariant statistics has received remarkable confirmation from many quite distinct tests .
why does this hold ?
it certainly is nt true for auditory or tactile signals .
i think there is one major and one minor reason for it .
the major one is that the world is viewed from a random viewpoint , so one can move closer or farther from any scene .
to first approximation , this scales the image ( though not exactly because nearer objects scale faster than distant ones ) .
the minor one is that most objects are opaque but have , by and large , parts or patterns on them and , in turn , belong to clusters of larger things .
this observation may be formulated as saying the world is not merely made up of objects but it is cluttered with them . the natural setting for
scale invariance is pass to the limit and model images as random functions @xmath112 of two real variables .
then the hypothesis is that a suitable function space supports a probability measure which is invariant under both translations and scalings @xmath123 , whose samples are ` natural images ' .
this hypothesis encounters , however , an infra - red and an ultra - violet catastrophe : + a ) the infra - red one is caused by larger and larger scale effects giving bigger and bigger positive and negative swings to a local value of @xmath117 . but these large scale effects are very low - frequency and this is solved by considering @xmath117 to be defined only modulo an unknown constant , i.e. it is a sample from a measure on a function space mod constants .
+ b ) the ultra - violet one is worse : there are more and more local oscillations of the signals at finer and finer scales and this contradicts lusin s theorem that an integrable function is continuous outside sets of arbitrarily small measure .
in fact , it is a theorem that _ there is no translation and scale invariant probability measure on the space of locally integrable functions mod constants_. this can be avoided by allowing images to be generalized functions .
in fact , the support can be as small as the intersection of all negative sobolev spaces @xmath124 .
to summarize what a good statistical theory of natural images should explain , we have scale - invariance as just described , kurtosis greater than 3 as described in section 2.1 and finally the right local properties : * hypothesis i * : : a theory of images is a translation and scale invariant probability measure on the space of generalized functions @xmath112 mod constants .
* hypothesis ii * : : for any filter @xmath125 with mean 0 , the marginal statistics of @xmath126 have kurtosis greater than 3 . *
hypothesis iii * : : the local statistics of images reflect the preferred local geometries , esp .
images of straight edges , but also curved edges , corners , bars , ` t - junctions ' and ` blobs ' as well as images without geometry , blank ` blue sky ' patches .
hypothesis iii is roughly the existence of what marr , thinking globally of the image called the _ primal sketch _ and what julesz , thinking locally of the elements of texture , referred to as _ textons_. by scale invariance , the local and global image should have the same elements . to quantify hypothesis iii ,
what is needed is a major effort at data mining .
specifically , the natural approach seems to be to take a small filter bank of zero mean local filters @xmath127 , a large data base of natural images @xmath128 leading to the sample of points in @xmath129 given by @xmath130 for all @xmath131 .
one seeks a good non - parametric fit to this dataset . but
hypothesis iii shows that this distribution will not be simple .
for example lee et al [ l - p - m ] have taken @xmath132 , @xmath133 a basis of zero mean filters with fixed @xmath134 support .
they then make a linear tranformation in @xmath135 normalizing the covariance of the data to @xmath136 ( ` whitening ' the data ) , and to investigate the outliers , map the data with norms in the upper 20% to @xmath137 by dividing by the norm .
the analysis reveals that the resulting data has asymptotic infinite density along a non - linear surface in @xmath137 ! this surface is constructed by starting with an ideal image , black and white on the two sides of a straight edge and forming a @xmath134 discrete image patch by integrating this ideal image over a tic - tac - toe board of square pixels .
as the angle of the edge and the offset of the pixels to the edge vary , the resulting patches form this surface .
this is the most concrete piece of evidence showing the complexity of local image statistics . are there models for these three hypotheses ?
we can satisfy the first hypothesis by the unique scale - invariant gaussian model , called the free field by physicists but its samples look like clouds and its marginals have kurtosis 3 , so neither the second nor third hypothesis is satisfied .
the next best approximation seems to be to use infinitely divisible measures , such as the model constructed by the author and b.gidas [ m - g ] , which we call _ random wavelet expansions _ : @xmath138 where @xmath139 is a poisson process in @xmath140 and @xmath141 are samples from an auxiliary levi measure , playing the role of individual random wavelet primitives .
but this model is based on adding primitives , as in a world of transparent objects , which causes the probability density functions of its marginal filter statistics to be smooth at 0 instead of having peaks there , i.e. the model does not produce enough ` blue sky ' patches with very low constrast .
a better approach are the random collage models , called _
dead leaves models _ by the french school : see [ l - m - h ] . here
the @xmath141 are assumed to have bounded support , the terms have a random depth and , instead of being simply added , each term occludes anything behind it with respect to depth .
this means @xmath112 equals the one @xmath141 which is in front of all the others whose support contains @xmath142 .
this theory has major troubles with both infra - red and ultra - violet limits but it does provide the best approximation to date of the empirical statistics of images .
it introduces explicitly the hidden variables describing the discrete objects in the image and allows one to model their preferred geometries .
crafting models of this type is not simply mathematically satisfying .
it is central to the main application of computer vision : object recognition .
when an object of interest is obscured in a cluttered badly lit scene , one needs a @xmath44-value for the hypothesis test is this fragment of stuff part of the sought - for object or an accidental conjunction of things occurring in generic images ? to get this @xmath44-value , one needs a null hypothesis , a theory of generic images .
-5 mm as we have seen in the last section , modeling images leads to objects and these objects have shape so we need stochastic models of shape , the ultimate non - linear sort of thing .
again it is natural to consider this in the continuum limit and consider a @xmath49-dimensional shape to be a subset of @xmath129 , e.g. a connected open subset with nice boundary @xmath115 .
it is very common in multiple images of objects like faces , animals , clothes , organs in your body , to find not identical shapes but warped versions .
how is this to be modeled ?
one can follow the ideas of the previous section and take a highly empirical approach , gathering huge databases of faces or kidneys .
this is probably the road to the best pattern recognition in the long run . but
another principle that grenander has always emphasized is to take advantage of the group of symmetries of the situation in this case , the group of all diffeomorphisms of @xmath129 .
he and miller and collaborators ( see [ gr - m ] ) were led to rediscover the point of view of arnold which we next describe .
let @xmath143 and @xmath144 be the volume - preserving subgroup .
we want to bypass issues of the exact degree of differentiability of these diffeomorphisms , but consider @xmath145 and @xmath144 as infinite dimensional riemannian manifolds .
let @xmath146 be a path in @xmath144 and define its length by : @xmath147 this length is nothing but the _ right_-invariant riemannian
metric : @xmath148 arnold s beautiful theorem is : * theorem * _ geodesics in @xmath144 are solutions of euler s equation : _
@xmath149 this result suggests using geodesics on suitable infinite dimensional manifolds to model optimal warps between similar shapes in images and using diffusion on these manifolds to craft stochastic models .
but we need to get rid of the volume - preserving restriction .
the weak metric used by arnold no longer works on the full @xmath145 and in [ c - r - m ] , christensen et al introduced : @xmath150 where @xmath105 is any vector field and @xmath151 is a fixed positive self - adjoint differential operator e.g. @xmath152 then a path @xmath153 in @xmath76 has both a velocity : @xmath154 and a _ momentum : _
@xmath155 ( so @xmath156 , @xmath157 the green s function of @xmath151 ) .
what is important here is that the momentum @xmath158 can be a generalized function , even when @xmath159 is smooth .
the generalization of arnold s theorem , first derived by vishik , states that geodesics are : @xmath160 this equation is a new kind of regularized compressible euler equation , called by marsden the template matching equation ( tme ) .
the left hand side is the derivative along the flow of the momentum , as a measure , and the right hand side is the force term .
a wonderful fact about this equation is that by making the momentum singular , we get very nice equations for geodesics on the @xmath145-homogeneous spaces : + ( a ) @xmath161 set of all @xmath30-tuples of distinct points in @xmath162 and + ( b ) @xmath163 set of all images of the unit ball under a diffeomorphism . + in the first case , we have @xmath164 where @xmath165 is the stabilizer of a specific set @xmath166 of @xmath30 distinct points . to get geodesics on @xmath167
, we look for ` particle solutions of the tme ' , i.e.@xmath168 where @xmath169 is a path in @xmath167 the geodesics on @xmath145 , which are perpendicular to all cosets @xmath170 , are then the geodesics on @xmath167 for the quotient metric : @xmath171 where @xmath172 . for these we get the interesting hamiltonian ode : @xmath173 which makes points traveling in the same direction attract each other and points going in opposite directions repel each other . this space leads to a non - linear version of the theory of landmark points and shape statistics of kendall [ sm ] and has been developed by younes [ yo ] .
a similar treatment can be made for the space of shapes @xmath174 , where @xmath175 is the stabilizer of the unit sphere .
geodesics on @xmath176 come from solutions of the tme for which @xmath177 is supported on the boundary of the shape and perpendicular to it .
even though the first of these spaces @xmath178 might seem to be quite a simple space , it seems to have a remarkable global geometry reflecting the many perceptual distinctions which we make when we recognize a similar shapes , e.g. a cell decomposition reflecting the different possible graphs which can occur as the ` medial axis ' of the shape .
this is an area in which i anticipate interesting results .
we can also use these riemannian structures to define brownian motion on @xmath179 and @xmath167 ( see [ d - g - m ] , [ yi ] ) . putting a random stopping time on this walk ,
we get probability measures on these spaces . to make the ideas more concrete ,
in figure 6 we show a simulation of the random walk on @xmath178 .
-5 mm the patterns which occur in nature s sensory signals are complex but allow mathematical modeling .
their study has gone through several phases . at first , ` off - the - shelf ' classical models ( e.g. linear gaussian models ) were adopted based only on intuition about the variability of the signals .
now , however , two things are happening : computers are large enough to allow massive data gathering to support fully non - parametric models .
and the issues raised by these models are driving the study of new areas of mathematics and the development of new algorithms for working with these models .
applications like general purpose speech recognizers and computer driven vehicles are likely in the foreseeable future .
perhaps the ultimate dream is a fully unsupervised learning machine which is given only signals from the world and which finds their statistically significant patterns with no assistance : something like a baby in its first 6 months . | is there a mathematical theory underlying intelligence ?
control theory addresses the output side , motor control , but the work of the last 30 years has made clear that perception is a matter of bayesian statistical inference , based on stochastic models of the signals delivered by our senses and the structures in the world producing them
. we will start by sketching the simplest such model , the hidden markov model for speech , and then go on illustrate the complications , mathematical issues and challenges that this has led to .
4.5 mm * keywords and phrases : * perception , speech , vision , bayesian , statistics , inference , markov . |
determining the chemical composition of galaxies is of fundamental importance for tracing back the history of evolution of galaxies .
in particular , the lmr has been widely studied in the local universe .
metallicities are tightly related with the luminosities of galaxies in such a way , that brighter systems have higher abundances ( lamareille et al .
furthermore , recent studies have also suggested that this relation extends to intermediate redshifts but displaced towards lower metallicities and higher luminosities ( kobulnicky et al .
2003 ) .
when studying galaxy evolution , stellar mass is a better parameter than luminosity .
however , because of the difficulties in obtaining stellar masses , most studies have used luminosity as a surrogate .
recently , though , tremonti et al . (
2004 ) have estimated the relation between metallicity and stellar mass in the local universe .
the authors found a strong correlation extended over 2 dex in stellar mass and a factor of 10 in metallicity . in this work ,
we study the evolution of the mmr and the lmr by employing numerical chemo - dynamical simulations which allow to describe the non - lineal growth of structure simultaneously with the enrichment of the interstellar medium in a cosmological framework .
we have run numerical simulations by using the chemical gadget-2 of scannapieco et al .
( 2005 ) . a @xmath0cdm cosmological model ( @xmath4=0.3 , @xmath0=0.7 , @xmath5=0.04 and @xmath6=100 @xmath7 km s@xmath8 mpc@xmath8 with @xmath7=0.7 ) was assumed , according to which galaxies formed by the hierarchical aggregation of substructures .
we have analysed two realizations of the power spectrum in a 10 mpc @xmath9 side box , initially resolved with @xmath10 ( s160 ) and @xmath11 ( s80 ) particles , corresponding to mass resolutions of @xmath12 m@xmath13 and @xmath14 m@xmath13 for the gas phase and @xmath15 m@xmath13 and @xmath16 m@xmath13 for dark matter respectively .
a salpeter initial mass function has been assumed with upper and lower limits of 40 m@xmath17 and 0.1 m@xmath17 , respectively .
the chemical model includes metal - dependent radiative cooling , star formation and chemical enrichment by supernovae ii and ia ( scannapieco et al . 2005 ) .
galactic objects were identified by applying an identification algorithm that combines the friends - of - friends technique and the contrast density criterium of white , efstathiou & frenk ( 1993 ) .
dynamical and chemical properties were estimated at the optical radius calculated accordingly to the standard definition as the radius which contains the 83% of the baryonic mass of the system ( tissera et al .
colours and magnitudes of galactic systems were calculated by resorting to population synthesis models ( see de rossi et al .
2006 in preparation ) .
our simulations predict a linear correlation between oxygen abundance and luminosity which is in good agreement with the observational results .
we have also found an evolution in the lmr in such a way that the slope increases and the zero point decreases with redshift consistently with the findings of kobulnicky & kewley ( 2004 ) , among others .
in particular , we decided to work with the i - band because it is less affected by extinction and can be more directly related with the underlying mass distributions .
our results indicate that at a given chemical abundance , galactic systems are @xmath18 3 dex brighter at @xmath19 compared to @xmath20 , with the larger evolution at fainter magnitudes .
futhermore , we have encountered a mean evolution in the chemical abundances of galactic systems of @xmath18 1.6 dex for brighter magnitudes and @xmath18 2.5 dex for faint ones , from @xmath19 to @xmath20 .
we have also analysed the mmr for simulated galactic systems , obtaining similar trends to those found by tremonti et al .
( 2004 ) in the sloan digital sky survey ( sdss ) but with a displacement of -0.25 dex in the zero point .
this last difference may be explained taking into account that the sdss explored only the central regions of galaxies which could lead to an overestimation of their metal content .
galactic abundances derived from simulations tend to increase with stellar mass which is also consistent with the observed behaviour .
however , we obtained an excess of metals in the lower mass end which could be due to the absence of supernovae energy feedback in our model .
we have determined a characteristic stellar mass at @xmath21 m@xmath13 where a change in the curvature of the mmr occurs .
this characteristic mass , which corresponds to an oxygen abundance of @xmath18 8.7 dex , has been obtained by estimating where the residuals of the linear fits depart systematically from zero .
it is important to note that this mass is similar to the characteristic mass derived from the sdss by tremonti et al .
( 2004 ) and gallazzi et al.(2005 ) .
in addition , we have found that the mmr exhibits the same general patterns from @xmath19 to @xmath20 , but with a displacement towards higher abundances as redshift decreases .
the characteristic stellar mass @xmath22 remains almost unchanged with time and only its corresponding chemical abundance evolves by 0.05 dex in the same redshift range .
the major departure from the local mmr occurs for smaller systems which increase their chemical content by @xmath18 0.10 dex . on the other hand ,
massive systems show less evolution with variations of @xmath18 0.05 from @xmath19 to @xmath20 . we have also studied the metallicity - optical velocity relation ( mvr ) finding a well defined correlation from @xmath19 to @xmath20 .
however , a higher level of evolution has been found in the mvr when compared to the mmr .
fast rotators show an enrichment of @xmath18 0.18 dex from @xmath19 to @xmath20 while at lower metallicities the variations are of @xmath18 0.28 dex .
this significant evolution of the mvr is a consequence of the increase of the mean density of the universe at high redshift , so that at a fix stellar mass , systems need to concentrate more in order to fullfill the contrast density criterium and , hence , galaxies reach higher velocities .
the larger evolution showed by the simulated mvr , when compared with the mmr , shows that the second is more fundamental in the sense that it is not strongly dependent on the cosmic epoch ( see tissera , de rossi & scannapieco 2005 for details ) . by analysing the merger trees of systems at @xmath20 , we have encountered that the features of the mmr can be traced back in time taking into account the mergers and interaction history of galaxies within the hierarchical aggregation picture .
more massive systems transform most of their gas into stars at high redshifts and experience important mergers . at lower redshifts
, these galactic objects are saturated of stars , so that , while their stellar mass largely increases in a merger event , their metallicity remains basically unchanged .
on the other hand , less massive systems form their stars in a more passive way or by rich - gas mergers leading to a more tight correlation between metallicity and stellar mass .
we have estimated the mmr , the lmr and the mvr correlations finding similars trends to those detected in observations .
these relations evolve with redshift with the major changes driven by less massive systems in consistency with the downsizing scenario .
gallazzi , a. , charlot , s. , brinchmann , j. , white , s. d. m. , tremonti , c. , 2005 , mnras , accepted ( astroph/0506539 ) kobulnicky , h. a. , willmer , c. n. a. , phillips , a. c. , koo , d. c. , faber , s. m. et al .
2003 , apj 599 , 1006 kobulnicki , h. a. , kewley , l. j. 2004 , apj 617 , 240 lamareille , f. , mouhcine , m. , contini , t. , lewis , i. , maddox , s. 2004 , mnras , 350 , 396 scannapieco , c. , tissera , p. b. , white , s. d. m. , springel , v. , 2005 , mnras , accepted ( astroph/0505440 ) tissera , p. b. , de rossi , m. e. , scannapieco , c. , 2005 , mnras , 364 , 38l tremonti , c. a. , heckman , t. m. , kauffmann , g. , brinchmann , j. , charlot , s. et al .
2004 , apj 613,898 white , s. d. m. , efstathiou , g. & frenk , c. s. , 1993 , mnras , 262 , 1023 | in this work , we study the luminosity - metallicity relation ( lmr ) and the stellar mass - metallicity relation ( mmr ) of galactic systems in a hierarhical clustering scenario .
we performed numerical hydrodynamical simulations with the chemical gadget-2 of scannapieco et al.(2005 ) in a @xmath0cdm universe .
we found that our simulated galactic systems reproduce the observed local lmr and its evolution in zero point and slope .
the simulated mmr is also in agreement with recent observational results . from the analysis of the evolution of the mmr
, we found a characteristic mass at @xmath1 which separates two galactic populations with different astrophysical properties .
more massive systems tend to have their stars formed at @xmath2 and show less evolution than smaller systems .
hence , this characteristic mass is determined by the formation of the structure in a hierarchical scenario .
our results also suggest the need for efficient supernova feedback .
1.0 cm 0.3 cm en este trabajo , estudiamos la relacin luminosidad - metalicidad ( lmr ) y la relacin masa estelar - metalicidad ( mmr ) de los sistemas galcticos en un modelo de agregacin jerrquica .
realizamos simulaciones numricas hidrodinmicas con el cdigo qumico gadget-2 de scannapieco et al .
( 2005 ) en un universo @xmath0cdm .
encontramos que nuestros sistemas galcticos simulados reproducen la lmr local y su evolucin en el punto cero y la pendiente .
la mmr simulada est tambin en acuerdo con resultados observacionales recientes . a partir del anlisis de la evolucin de la mmr , hallamos una masa caracterstica en @xmath1 , la cual separa dos poblaciones galcticas con diferentes propiedades astrofsicas .
los sistemas ms masivos tienden a formar sus estrellas a @xmath3 y muestran menor evolucin que los sistemas pequeos .
entonces , esta masa caracterstica es determinada por la formacin de la estructura en un universo jerrquico .
nuestros resultados sugieren tambin la necesidad de un importante _ feedback _ por supernovas . |
sample covariance matrices are fundamental objects in modern multivariate statistics . in the classical setting @xcite , for an @xmath8 matrix @xmath1 , people focus on the asymptotic properties of @xmath9 when @xmath10 is fixed and @xmath11 goes to infinity . in this case
the central limit theorems and laws of large number can be applied to the statistical inference procedure .
however , the advance of technology has led to high dimensional data where @xmath10 is comparable to or even larger than @xmath11 @xcite .
this high dimensionality can not be handled with the classical multivariate statistical theory . in modern statistical study of sample covariance matrices
@xcite , the largest eigenvalues have been playing important roles in analyzing the covariance matrices .
on the one hand , it is of great interest to the principal component analysis ( pca ) @xcite , which is a standard technique for dimensionality reduction and provides a way to identify patterns from real data . as an example , in many applications @xcite the largest eigenvalue explains most of the variance of the data . on the other hand ,
the largest eigenvalue is commonly used in hypothesis testing , such as in the well - known roy s largest root test @xcite .
roy s largest root test has been used not only in detecting signals from the noisy data , but also in testing the group differences in manova .
therefore the distribution of the largest eigenvalue is important both in finding the second order asymptotics for the estimate in pca and in computing the power of test in hypothesis testing . for a detailed review
, one can refer to @xcite . in this paper
, we study an @xmath8 matrix @xmath1 and the covariance matrix @xmath12 in the regime where @xmath13 are proportional to each other . in the last decades
, random matrix theory has been proved to be one of the powerful tools in dealing with large dimensional covariance matrices . in the framework of random matrix theory
, people managed to show that the empirical spectral distribution ( esd ) of @xmath12 converges to the marchenko - pastur ( mp ) law @xcite , and the distribution of the largest eigenvalue converges to the tracy - widom distribution under a proper @xmath14 scaling .
it is notable that these results are universal , i.e. they are independent of the detailed distribution of the entries of @xmath1 .
the results were first proved for wishart matrix ( i.e. the covariance matrix obtained from a data matrix @xmath1 consisting of i.i.d centered real or complex gaussian random entries ) @xcite ; later they were proved for matrices @xmath1 with entries satisfying arbitrary sub - expoenetial distribution @xcite . in this paper
, we again study the edge universality for the largest eigenvalues of sample covariance matrices and we want to relax the assumptions on the distribution of the entries in @xcite .
more specifically , we prove a necessary and sufficient condition for the edge universality of covariance matrices .
this was a conjecture proposed by pillai and yin @xcite , and its counterpart for wigner matrices has been proved by lee and yin in @xcite .
[ def_convariance ] let @xmath15 be an @xmath16 data matrix with independent centered entries @xmath17 , @xmath18 and @xmath19 .
then @xmath0 defines an @xmath20 covariance matrix .
moreover , we define the following two conditions for the matrix entries : @xmath21 @xmath22 are i.i.d random variables with @xmath23 @xmath24 @xmath25 satisfies the following tail condition @xmath26 in this paper , we denote the eigenvalues of @xmath0 as @xmath27 .
we will also use @xmath28 and its eigenvalues @xmath29 in the proof .
since @xmath28 and @xmath12 share the same nonzero eigenvalues , we will simply use @xmath30 , @xmath31 , to denote the @xmath32-th eigenvalue of both @xmath28 and @xmath12 without causing any confusion .
we are interested in the regime @xmath33 in this paper , we often abuse the notation a little bit and write @xmath34 as @xmath35 .
we also define the soft edge for the largest eigenvalue @xmath36 we are interested in the distribution of @xmath37 and the main result of this paper is the following theorem . for the precise statement
, we refer the reader to theorem [ main_thm ] .
[ main_thm_simple ] for any matrix @xmath1 satisfying condition @xmath38 , @xmath39 converges weakly to the tracy - widom distribution if and only if @xmath1 satisfies condition @xmath40 .
theorem [ main_thm_simple ] provides a simple necessary and sufficient condition for the edge universality of covariance matrices without assuming any other properties of matrix entries .
recall that in @xcite , the authors proved that @xmath41 a.s . if the fourth moment exists . furthermore in @xcite
, the finite fourth moment condition is proved to be necessary for the almost sure convergence of @xmath37 .
our theorem , however , shows that the existence of finite fourth moment is not necessary for the tracy - widom fluctuation .
for instance , if the entries of @xmath1 have the following probability density function @xmath42 then the fourth moment does not exist , while theorem [ main_thm_simple ] still applies .
thus in this case @xmath37 does not converge to @xmath43 a.s . , but @xmath39 still converges weakly to the tracy - widom distribution . on the other hand , under the condition @xmath40 , silverstein proved that @xmath41 in probability @xcite .
so theorem [ main_thm_simple ] can be regarded as an improvement of that result .
the main idea of our proof basically follows from that for wigner matrices in @xcite . in @xcite , the local laws and the edge universality for covariance matrices have been proved by assuming a subexponential decay of the @xmath44 entries . using the methods in @xcite , we extend these results to the family of covariance matrices with entries having size @xmath45 for some @xmath46 ( i.e. @xmath1 has support bounded by @xmath47 , see definition [ defn_support ] ) .
a key observation is that if we introduce a cutoff " on the matrix elements at @xmath48 , then the matrix with the cutoff can well approximate the original matrix in terms of the largest singular value if and only if the condition ( [ introductioneq ] ) holds @xcite .
thus a major part of this paper is devoted to extend the
small " support @xmath47 , @xmath46 , results to the large " support @xmath48 case .
this can be accomplished with a green function comparison method , which has been used successfully in proving the universality of wigner and covariance matrices @xcite .
more specifically , we will use a lindeberg replacement strategy for the comparison .
different from @xcite , where the lindeberg replacement is performed by changing columns of @xmath1 , we need to change @xmath1 entry by entry in our proof .
this paper is organized as follows . in section [ main_result ] , we give the main theorem and introduce some notations and tools that will be used in our proof . in section [ sec_cutoff ] , we prove the main result using the cutoff argument . in sections [ tools ] and [ comparison ] , we prove the key lemmas and theorems that are used in section [ sec_cutoff ] . in particular , the green function comparison is performed in section [ comparison ] . in appendix
[ appendix1 ] , we prove the local laws and edge universality for sample covariance matrices with small support @xmath47 for some @xmath46 . for definiteness , in this paper
we focus on the real case , i.e. all the entries @xmath44 are real .
however , it is clear that our proof can be applied to the complex case after minor modifications if we assume in addition that @xmath49 and @xmath50 are independent centered random variables with same variance .
also in this paper , we do not consider the edge universality at the hard edge ( i.e. , for the smallest singular values ) .
so far , there is no conjecture about the necessary and sufficient condition for the edge universality at the hard edge .
* conventions . *
all quantities that are not explicitly constant may depend on @xmath11 , and we usually omit @xmath11 from our notations .
we use @xmath51 to denote a generic large positive constant , whose value may change from one line to the next . similarly , we use @xmath52 or @xmath53 to denote a generic small positive constant . for two quantities @xmath54 and @xmath55 depending on @xmath11 , the notation @xmath56 means that @xmath57 for some positive constant @xmath58 , and @xmath59 means that @xmath60 for some positive constants @xmath61 as @xmath62 .
we also use the notation @xmath63 if @xmath56 and @xmath64 . for a matrix @xmath65
, we use @xmath66 to denote its operator norm and @xmath67 the hilbert - schmidt norm ; for a vector @xmath68 , @xmath69 stands for the euclidean norm , while @xmath70 stands for the @xmath71-norm . in this paper
, we usually write an @xmath72 identity matrix @xmath73 as @xmath74 or @xmath75 when there is no confusion about the dimension .
if two random variables @xmath1 and @xmath76 have the same distribution , we write @xmath77 .
in this section , we introduce some notations that will be used in the proof . recall that the empirical spectral distribution ( esd ) of an @xmath78 symmetric matrix @xmath79 is defined as @xmath80 we define the green functions for @xmath12 and @xmath28 as @xmath81 the stieltjes transform of the esd of @xmath82 is given by @xmath83 similarly , we can also define @xmath84 .
since the nonzero eigenvalues of @xmath28 and @xmath12 are identical and @xmath12 has @xmath85 more ( or @xmath86 less ) zero eigenvalues , we have @xmath87 and @xmath88 for @xmath1 in definition [ def_convariance ] , under the assumption ( [ notations ] ) , the esd of @xmath0 converges to the marchenko - pastur ( mp ) law as @xmath62 @xcite : @xmath89 then @xmath90 converges to the stieltjes transform of the mp law @xmath91 , which satisfies the following self - consistent equation ( see ( 1.2 ) of @xcite ) @xmath92 and has the closed form expression @xmath93 from ( [ barm ] ) ( see also ( 1.5 ) of @xcite ) , it is easy to see that @xmath94 converges to @xmath95 as @xmath62 , where @xmath96 moreover , it is easy to verify that @xmath97 satisfies the self - consistent equation ( see also ( 1.4 ) of @xcite and ( 1.12 ) of @xcite ) @xmath98 correspondingly , we denote the probability density function for the limiting esd of @xmath99 as @xmath100 we define the distance of @xmath101 to the right edge @xmath102 ( see ( [ notations1 ] ) and ( [ mp_density ] ) ) as @xmath103 let @xmath104 we define two sets as @xmath105 where @xmath106 is a constant .
note that we do not include the hard edge ( i.e. , the smallest singular value ) in the definition of @xmath107 , since we are only interested in the largest singular values in this paper . also with @xmath108 in ( [ mp_density ] ) , it is easy to see that @xmath109 on @xmath107 for some constant @xmath110 depending on @xmath35 only . in the following lemma , we summarize some basic behaviors of @xmath111 and @xmath112 , which can be proved with direct calculations ( or see lemma 2.3 in @xcite ) .
[ lem_mbehavior ] the following statements hold .
* we have @xmath113.\label{squareroot}\ ] ] * for @xmath114 , we have @xmath115 and @xmath116 [ high_prob ] we say that an @xmath11-dependent event @xmath117 holds with @xmath118-high probability ( @xmath119 ) if there exists constant @xmath58 independent of @xmath11 , such that @xmath120 for all sufficiently large @xmath11 , where @xmath121 is defined in ( [ def_phi ] ) . for simplicity ,
when @xmath122 , we just say high probability .
note that if @xmath117 holds with @xmath118-high probability , then @xmath123 for any @xmath124 . [ defn_support ] a family of matrices @xmath125 are said to satisfy the bounded support condition with @xmath126 if @xmath127 : @xmath128 for some @xmath129 . here
@xmath126 often depends on @xmath11 and satisfies @xmath130 for some small positive constant @xmath131 .
whenever ( [ eq_support ] ) holds , we will say that @xmath44 has support @xmath132 .
note that the gaussian distribution satisfies the condition ( [ eq_support ] ) with @xmath133 for any @xmath134 .
we also remark that by definition [ high_prob ] , the event @xmath135 in ( [ eq_support ] ) holds with @xmath118-high probability for any positive constant @xmath119 . for this reason ,
the bad event @xmath136 is negligible during the proof , and we will not consider the case it happens . before concluding this section , we introduce a self - adjoint linearization trick , which has been proved to be useful in studying the local laws of non - hermitian matrices @xcite .
it will simplify our proof in many ways .
we define the following @xmath137 block matrix , which is a linear function of @xmath1 .
[ def_linearhg]for @xmath138 , we define the @xmath137 self - adjoint matrices @xmath139 and @xmath140 for @xmath79 and @xmath141 , we define the index sets @xmath142 when we use @xmath79 or @xmath141 , we will consistently use the latin letters @xmath143 , greek letters @xmath144 , and @xmath145 .
then we label the indices of the matrices according to @xmath146 for @xmath147 and @xmath148 , we introduce the notation @xmath149 and @xmath150 . for an @xmath151 matrix @xmath65 , we denote the @xmath152 matrices @xmath153}$ ] as @xmath154}=\left ( { \begin{array}{*{20}c } { a_{ij } } & { a_{i\bar j } } \\ { a_{\bar i j } } & { a_{\bar i\bar j } } \\ \end{array } } \right ) , \ \
1\le i , j \le \min\{n , m\}.\ ] ] we shall call @xmath153}$ ] a diagonal group if @xmath155 , and an off - diagonal group otherwise . it is easy to verify that the eigenvalues @xmath156 of @xmath79 are @xmath157 and @xmath158 where we use the notations @xmath159 and @xmath160 . in particular , the largest eigenvalue @xmath161 can be studied through the largest eigenvalue @xmath162 of @xmath79 .
on the other hand , by schur complement formula , we have @xmath163 for @xmath164 defined in ( [ def_green ] ) .
thus a control of @xmath141 yields directly a control of @xmath165 and @xmath166 .
moreover , we have @xmath167 finally we introduce the spectral decomposition of @xmath141 .
let @xmath168 be the singular decomposition of @xmath1 , where @xmath169 and @xmath170 and @xmath171 are orthonormal bases of @xmath172 and @xmath173 , respectively . then by ( [ green2 ] )
, we have for @xmath174 and @xmath175 , @xmath176 the main result of this paper is the following theorem [ main_thm ] .
it establishes the necessary and sufficient condition for the edge universality of the covariance matrices @xmath0 at the soft edge ( i.e. , for the largest eigenvalues @xmath177 ) .
[ main_thm ] let @xmath178 and @xmath1 be @xmath16 i.i.d matrices satisfying condition @xmath179 ( see definition [ def_convariance ] ) . moreover ,
suppose the entries of @xmath180 satisfy the gaussian distribution .
* * sufficient condition * : if @xmath1 satisfies condition @xmath181 , then for any @xmath182 , @xmath183 as @xmath62 , where @xmath184 is the law for @xmath180
. in particular , this implies that as @xmath62 , the distribution of @xmath37 converges weakly to the tracy - widom distribution .
* * necessary condition * : if condition @xmath40 does not hold for @xmath1 , then the distribution of @xmath37 does not converge to the tracy - widom distribution . furthermore , for any @xmath185
, we have @xmath186 in @xcite , the tracy - widom law was proved to hold for the largest eigenvalues of covariance matrices of the form @xmath187 , where @xmath188 is a positive definite symmetric matrix .
this was proved under the assumption that the entries of @xmath1 have a distribution with sub - exponential decay . in @xcite ,
the edge universality for the largest eigenvalues of covariance matrices @xmath189 was established for much more general @xmath190 ( which is not necessarily to be a square matrix ) .
we conjecture that @xmath191 is the necessary and sufficient condition for the edge universality of @xmath192 when @xmath190 is diagonal and satisfies some mild assumptions . in this case
, the generalized local mp law can be established ( see lemma [ lem_small ] ) and all our proofs should work . on the other hand ,
if @xmath190 is not diagonal , we need the so - called anisotropic local law " to establish the local mp law for @xmath192 .
then the sub - exponential decay assumption is crucial for this purpose , and so far we have no idea how to relax it essentially .
the main idea of our proof is similar to the one for wigner matrices in @xcite . for the sufficient condition
, we first prove the edge universality for covariance matrices satisfying conditions @xmath193 with small support @xmath194 .
then we extend the result to the covariance matrices with larger support @xmath195 .
as we will see in the proof , the large support case leads to the conclusion ( [ sufficient ] ) .
for the small support case , we will follow the proofs in @xcite .
for the large support case , we use a green function comparison method as in @xcite : we expand the green functions with large supports in terms of green functions with small supports ( see section [ comparison ] ) using resolvent expansions and estimate the related errors .
now we summarize the relevant results in the small support case . the proof is put in appendix [ appendix1 ] .
[ classical locations of eigenvalues ] the classical location @xmath196 of the @xmath32-th eigenvalue of @xmath0 is defined as @xmath197 where @xmath198 is defined in ( [ mp_density ] ) .
we define the deterministic limit @xmath199 of the green function in ( [ green2 ] ) as @xmath200 then we have the following form of the local mp law . [ lem_small ]
let @xmath1 be a matrix satisfying conditions @xmath38 and @xmath127 with @xmath201 for some @xmath202 .
then there exists constants @xmath203 , @xmath204 and @xmath205 such that for @xmath206 in ( [ eqn_defg ] ) , the following results hold with @xmath207-high probability : 1 . *
( local mp law ) * : @xmath208 @xmath209 2 . *
( bound on @xmath210 ) * : @xmath211 3 . * ( delocalization ) * : @xmath212 4 . *
( rigidity of eigenvalue ) * : if @xmath201 for some constant @xmath213 , @xmath214 where @xmath30 denotes the @xmath32-th eigenvalue of @xmath12 .
furthermore , we have the following result for the edge universality .
suppose @xmath215 and @xmath216 are two matrices satisfying conditions @xmath38 and @xmath127 with support @xmath201 for some @xmath217 .
then there exist constants @xmath218 such that , for any @xmath219 , we have @xmath220 where @xmath221 and @xmath222 denote the laws of @xmath216 and @xmath215 respectively .
[ rigid_multi ] ( i ) if @xmath223 , we can prove the following stronger results about the rigidity of eigenvalues , e.g. see ( * ? ? ?
* theorem 2.13 ) and ( * ? ? ?
* theorem 3.3 ) : for @xmath224 , there exist constants @xmath203 and @xmath225 such that the following results hold with @xmath207-high probability , @xmath226 where @xmath227 is defined as @xmath228 when @xmath229 , the smallest eigenvalue of @xmath12 is close to zero and the above result does not hold for @xmath32 near @xmath230 .
however , we do not consider the hard - edge ( i.e. the smallest eigenvalues ) in this paper and ( [ sec ] ) is sufficient for our use .
\(ii ) as in @xcite , we can extend ( [ edddd ] ) to finite correlation functions of largest eigenvalues .
for example , we have @xmath231 for any fixed @xmath232 and sufficiently large @xmath11 . as a result , ( [ sufficient ] ) can also be extended to finite correlation functions of largest eigenvalues .
the proof of ( [ edddd_ext ] ) is similar to that of ( [ edddd ] ) by using a general form of the green function comparison method . for any matrix @xmath1 satisfying conditions @xmath38 and @xmath233 , we can construct a matrix @xmath234 which is close " to @xmath1 with high probability and satisfies conditions @xmath193 with @xmath201 for some @xmath202 , and @xmath235 : @xmath236 for some constant @xmath237 .
we refer the reader to section [ sec_cutoff ] for the details .
then the following theorems on matrices with large support and satisfying condition @xmath238 play key roles in proving ( [ sufficient ] ) .
[ thm_largerigidity ] suppose @xmath1 satisfies conditions @xmath38 , @xmath239 with @xmath240 for some @xmath241 , and @xmath242 .
then there exists constant @xmath243 , depending on @xmath244 and @xmath245 , such that with high probability we have @xmath246 for sufficiently large @xmath11 , where @xmath247 is defined in ( [ sset1 ] ) .
moreover , this result implies that with high probability the following rigidity result holds for some @xmath248 : @xmath249 and @xmath250 where @xmath251 [ thm_largebound ] let @xmath1 be a matrix satisfying the conditions in theorem [ thm_largerigidity ] .
then for any @xmath252 and @xmath253 , we have the following bound @xmath254 for some large enough constant @xmath243 .
[ rigid_reduction ] the rigidity result ( [ p1p1p1p1p1 ] ) shows that to prove the sufficient part of theorem [ main_thm ] , we only need to consider @xmath255 in ( [ sufficient ] ) .
otherwise , ( [ p1p1p1p1p1 ] ) already gives the desired result .
the basic idea of proving theorems [ thm_largerigidity ] and [ thm_largebound ] is similar to the one in @xcite .
first , we want to establish a connection between the matrices in theorems [ thm_largerigidity]-[thm_largebound ] and the matrices with small supports as in lemma [ lem_small ] .
the following lemma gives a way to decrease the support of matrices without changing the first four moments .
its proof can be found in ( * ? ? ?
* lemma 5.1 ) . [ lem_decrease ]
suppose @xmath15 satisfies conditions @xmath38 , @xmath127 with @xmath240 for some @xmath202 , and @xmath242 .
then there exists another matrix @xmath256 , such that @xmath257 satisfies the bounded support condition @xmath127 with @xmath258 , and the first four moments of the entries of @xmath1 and @xmath257 match , i.e. @xmath259 by lemma [ lem_small ] , @xmath260 has the properties that we want to prove for @xmath1 .
then we will use the green function comparison method to prove that @xmath1 has the same properties , since @xmath1 and @xmath260 have the same first four moments . more specifically ,
we want to prove the following lemma .
[ lem_comparison ] let @xmath1 and @xmath257 be two matrices as in lemma [ lem_decrease ] .
then there exist constants @xmath218 such that , for any @xmath261 we have @xmath262 where @xmath263 and @xmath264 are the laws for @xmath1 and @xmath257 , respectively .
one may notice the difference between this lemma and ( [ edddd ] ) : in lemma [ lem_comparison ] , @xmath1 and @xmath257 have different sizes for the support .
in this section , we prove theorem [ main_thm ] using the results in section [ sec_tools ] .
we begin by proving the necessary condition .
assume that @xmath265 .
then we can find a constant @xmath266 and a sequence @xmath267 s.t .
@xmath268 as @xmath269 and @xmath270 denote @xmath271 and consider the event @xmath272 we first show that @xmath273 when @xmath274 holds .
let @xmath275 such that @xmath276 since @xmath277 , we have @xmath278 where in the last step we used that @xmath279 .
now we prove that @xmath280 .
we choose @xmath281 .
define another event @xmath282 with the choice @xmath283 , we have that @xmath284 for some constant @xmath285 depending only on @xmath286 , @xmath35 and @xmath190 . since @xmath287 for some @xmath288 independent of @xmath11 , the above inequality shows that @xmath289 .
suppose that @xmath290 holds with @xmath291 for some @xmath292 .
by markov inequality , we have @xmath293 since the events @xmath294 are independent of @xmath290 , taking union over all @xmath295 we get that @xmath296 this shows that @xmath280 and concludes the proof .
given the matrix @xmath1 satisfying conditions @xmath38 and @xmath233 , as remarked above theorem [ thm_largerigidity ] , we first construct a matrix @xmath234 that satisfies conditions @xmath38 , @xmath127 with @xmath240 for some @xmath202 , and @xmath235 . for any fixed @xmath131 , define @xmath297.\ ] ] by ( [ introductioneq ] ) and integration by parts , we
have for any @xmath298 and large enough @xmath11 , @xmath299 let @xmath300 be the distribution density of @xmath25 .
then we define independent random variables @xmath301 , @xmath302 , @xmath303 , @xmath304 and @xmath305 in the following ways : * @xmath301 has distribution density @xmath306 , where @xmath307 * @xmath302 has distribution density @xmath308 , where @xmath309 * @xmath303 is a bernoulli 0 - 1 random variable with @xmath310 and @xmath311 .
let @xmath312 , @xmath313 and @xmath314 be random matrices such that @xmath315 , @xmath316 and @xmath317 . by ( [ def1 ] ) , ( [ def2 ] ) and the fact that @xmath318 is bernoulli , it is easy to check that for independent @xmath312 , @xmath313 and @xmath314 , @xmath319 where by ( [ bbbounds ] ) , we have @xmath320 therefore , if we define the matrix @xmath76 by @xmath321 , we have @xmath322 for some constant @xmath129 depending on @xmath53 and @xmath35 . then from the representation ( [ linearize_block ] ) , it is easy to see that @xmath323 for some constant @xmath129 .
hence the deterministic part @xmath76 in ( [ t3 ] ) is negligible due to remark [ rigid_reduction ] and the fact that @xmath324 is much smaller than @xmath48 . by ( [ introductioneq ] ) and integration by parts , it is easy to check that @xmath325 note that @xmath326 is a matrix that satisfies the assumptions for @xmath1 in theorems [ thm_largerigidity ] and [ thm_largebound ]
. then by lemma [ lem_decrease ] , we can find another matrix @xmath327 such that @xmath234 and @xmath327 satisfy the assumptions of lemma [ lem_comparison ] , and hence @xmath328 note that by lemma [ lem_small ] , @xmath329 satisfies the desired edge universality result .
together with the fact @xmath330 , we find that there exists constants @xmath331 such that for any @xmath219 , @xmath332 where @xmath333 is the law for @xmath312 and @xmath184 is the law for a gaussian covariance matrix .
now we write the first two terms in the right hand side of ( [ t3 ] ) as @xmath334 where @xmath335 it remains to deal with the @xmath336 term , and we call the corresponding matrix as @xmath337 . note that @xmath318 is independent of @xmath338 and @xmath339 .
we first introduce a cutoff on matrix @xmath314 as @xmath340 , where @xmath341 if we regard the matrix @xmath314 as a sequence @xmath342 of @xmath343 i.i.d bernoulli random variables , by the large deviation estimate we have that @xmath344 for some constant @xmath129 when @xmath11 is sufficiently large
. suppose the number @xmath232 of the nonzero elements in @xmath314 is given , then it is easy to check that @xmath345 combining the above estimates , we get that @xmath346 .
on the other hand , by condition ( [ introductioneq ] ) , we have @xmath347 for any fixed constant @xmath348
. therefore we can introduce the following matrix @xmath349 as @xmath350 where from the previous discussions , we see that @xmath351 .
therefore we only need to study the largest singular value of @xmath352 where @xmath353 and the rank of @xmath349 is less than @xmath354 .
we want to prove that @xmath355 where @xmath356 and @xmath357 .
recall that @xmath358 is independent of @xmath312 , so the position of the nonzero elements of @xmath349 is independent of @xmath312 . by symmetry
, we can assume the @xmath182 nonzero entries of @xmath349 are exactly @xmath359 now we define the matrices @xmath360 then we have the decomposition @xmath361 where @xmath362 is a @xmath363 diagonal matrix @xmath364 and @xmath365 is an @xmath366 matrix such that @xmath367 with the identity @xmath368 and lemma 6.1 of @xcite , if @xmath369 , then @xmath370 is an eigenvalue of @xmath371 if and only if @xmath372 where @xmath373 define @xmath374 for @xmath375 , it has the following @xmath152 blocks ( recall the definition ( [ aij_group ] ) ) : @xmath376 } \left ( { \begin{array}{*{20}c } { 1 } & { 1 } \\ { 1 } & { -1 } \\ \end{array } } \right ) + \delta_{ij}\left ( { \begin{array}{*{20}c } { ( \gamma e_{ii})^{-1 } } & { 0 } \\ { 0 } & { -(\gamma e_{ii})^{-1 } } \\
\end{array } } \right ) , \ \
1\le i \le s .\ ] ] now let @xmath377 we claim that for all @xmath378 , @xmath379 if ( [ suff_claim ] ) holds , then @xmath370 is not an eigenvalue of @xmath371 with probability @xmath380 .
denote the largest singular value of @xmath381 as @xmath382 , then by definition @xmath383 and @xmath384 . with the continuity of @xmath385 with respect to @xmath386 and the fact that @xmath387 , we find that @xmath388 with probability @xmath380 , i.e. we have proved ( [ eq_claim ] ) .
combining ( [ const_err ] ) , ( [ univ_small ] ) and ( [ eq_claim ] ) , we conclude ( [ sufficient ] ) .
it remains to prove the claim ( [ suff_claim ] ) .
choose @xmath389 and note that @xmath390 has support bounded by @xmath48 .
then by ( [ diagonal ] ) and ( [ squarerootbehavior ] ) , we have with high probability @xmath391 for the off - diagonal terms , we use ( [ weak_off ] ) , ( [ squarerootbehavior ] ) and markov inequality to conclude that @xmath392 holds with probability @xmath393 .
as pointed out in remark [ rigid_multi ] , we can extend ( [ univ_small ] ) to finite correlation functions of largest eigenvalues . since the largest eigenvalues in the gaussian case are separated in the scale @xmath394 , we conclude that @xmath395 on the other hand , the rigidity result ( [ p1p1p1p1p1 ] ) gives that with high probability , @xmath396 ) , ( [ rigid_estimate ] ) and the rigidity estimate ( [ p1p1p1p1p1 ] ) , we can get with probability @xmath380 that @xmath397 for instance , for @xmath398 , small @xmath399 and large enough @xmath58 , we have with probability @xmath380 that @xmath400 where in the first step we used ( [ spectral1 ] ) , in the second step ( [ delocal ] ) , in the third step @xmath401 for @xmath402 , in the fourth step ( [ repulsion_estimate ] ) , and in the last step the rigidity estimate ( [ p1p1p1p1p1 ] ) . for all other choices of @xmath403 and @xmath404 , we can prove the estimate ( [ realcomplex ] ) in a similar way .
now by ( [ realcomplex ] ) , we see that ( [ bound1 ] ) and ( [ bound2 ] ) still hold if we replace @xmath405 by @xmath406 and double the right hand sides . then using @xmath407 and ( [ q_group ] ) , we get that for any @xmath408 , @xmath409 and @xmath410 hold with probability @xmath380 .
thus @xmath411 is diagonally dominant with probability @xmath380 , which proves the claim ( [ suff_claim ] ) .
with lemma [ lem_decrease ] , given @xmath1 satisfying conditions @xmath38 , @xmath127 with @xmath240 for some @xmath202 , and @xmath242 , we construct a matrix @xmath257 that has support bounded by @xmath412 and has the same first four moments as @xmath1 .
moreover , @xmath260 has the desired edge universality according to lemma [ lem_small ] .
now we prove lemma [ lem_comparison ] , which shows that , similar as @xmath260 , @xmath1 also satisfies the edge universality .
our goal is to write the distribution of the largest eigenvalue in terms of a cutoff function depending on the green functions .
then it is natural to use the green function comparison result to conclude the proof .
let @xmath413 denote the number of eigenvalues of @xmath99 in @xmath414 $ ] ; similarly we define @xmath415 for @xmath416 . hence to quantify the distribution of @xmath37 , it is equivalent to use @xmath417 . recall remark [ rigid_reduction ]
, we can assume that the parameter @xmath182 satisfies @xmath418 and for any @xmath419 define @xmath420}$ ] to be the characteristic function of the interval @xmath421 $ ] .
for any @xmath422 , we define @xmath423 to be an approximate delta function on scale @xmath424 .
note that under the above definitions , we have @xmath425 and @xmath426 for any @xmath427 .
let @xmath132 be a smooth symmetric cutoff function such that @xmath428 and we assume that @xmath429 is decreasing when @xmath430 .
then the following lemma [ lem_cutoff2 ] provides a way to approximate @xmath431 with a function depending on green functions only .
[ lem_cutoff2 ] fix @xmath131 and let @xmath432 , @xmath433 .
suppose @xmath1 satisfies conditions @xmath434 @xmath127 with @xmath240 for some @xmath202 and @xmath242 .
then for all @xmath349 such that @xmath435 where the constant @xmath436 is as in ( [ p1p1p1p1p1 ] ) , ( [ pppppp ] ) , ( [ ps ] ) and ( [ eu ] ) , we have @xmath437 for some constant @xmath129 . [ lem_compdiffsupport ] suppose @xmath1 and @xmath257 satisfy the assumptions in lemma [ lem_decrease ] , and @xmath438 is a function whose derivatives satisfy @xmath439 for some constant @xmath440 .
then for any sufficiently small @xmath441 and any real numbers @xmath442 we have @xmath443 and @xmath444 for some constant @xmath445 , where @xmath446 is the stieltjes transform of the esd of @xmath447 .
now we conclude the proof of lemma [ lem_comparison ] .
recall that we only consider @xmath182 satisfying ( [ ps ] ) , so it suffices to assume @xmath448 throughout the following .
then by lemma [ lem_compdiffsupport ] and ( [ 512 ] ) , there exists @xmath298 such that @xmath449 for the choice @xmath433 , we also have @xmath450 .
thus we can apply lemma [ lem_cutoff2 ] to get @xmath451 combining ( [ 1 ] ) and ( [ 2 ] ) and using lemma [ lem_cutoff2 ] again , we get @xmath452 if we choose @xmath453 , then ( [ almostlast ] ) implies that @xmath454 this proves one part of ( [ eq_edgeuniv ] ) .
for the other inequality , the proof is similar .
[ lem_comgreenfunction ] let @xmath1 and @xmath257 satisfy the assumptions of lemma [ lem_decrease ] .
for @xmath455 in ( [ sset1 ] ) with large enough @xmath58 , if there exist deterministic quantities @xmath456 and @xmath457 , such that @xmath458 hold with @xmath207-high probability for some @xmath225 , then for any @xmath459 with @xmath460 , we have @xmath461 by lemma [ lem_decrease ] , @xmath257 has support bounded by @xmath412 . by ( [ squarerootbehavior ] )
, we have @xmath462 thus ( [ mpbounds ] ) and ( [ diagonal ] ) show that we can choose @xmath463 for some large enough @xmath464 such that ( [ keybounds ] ) holds with @xmath207-high probability . then using markov inequality and ( [ keyeyeyey ] )
, we get that for sufficiently large @xmath243 , @xmath465 for some constants @xmath466 . here
we use @xmath467 , the trivial bound @xmath468 on the bad event with probability @xmath469 , and the fact @xmath470 .
this proves ( [ newmpbounds ] ) , from which we can derive the rigidity results ( [ p1p1p1p1p1 ] ) and ( [ pppppp ] ) using the arguments in section 8 of @xcite , section 5 of @xcite or lemma 8.1 of @xcite . note that we can repeat the arguments in section 5 of @xcite almost verbartim except for some notations and coefficients . to prove lemmas [ lem_compdiffsupport ] and [ lem_comgreenfunction ]
, we will use the green function comparison method . more specifically
, we will apply the lindeberg replacement strategy to @xmath141 in ( [ eqn_defg ] ) .
let @xmath471 and @xmath472 be two matrices as in lemmas [ lem_compdiffsupport ] and [ lem_comgreenfunction ] . define a bijective ordering map @xmath473 on the index set of @xmath1 as @xmath474 for any @xmath475 , we define the matrix @xmath476 such that @xmath477 if @xmath478 , and @xmath479 otherwise .
note that @xmath480 and @xmath481 , and @xmath482 satisfies the bounded support condition with @xmath483 for all @xmath484 .
correspondingly , we define @xmath485 note that @xmath486 and @xmath487 differ only at @xmath488 and @xmath489 elements , where @xmath490 .
then we define the @xmath137 matrices @xmath365 and @xmath491 by @xmath492 so that @xmath486 and @xmath487 can be written as @xmath493 for some @xmath137 matrix @xmath494 satisfying @xmath495 . for simplicity , we denote the green functions @xmath496 by the above definitions , we can write @xmath497 thus we can expand @xmath498 using the resolvent expansion till the order of @xmath499 @xmath500 on the other hand we can expand @xmath501 in terms of @xmath498 , @xmath502 we have similar expansions for @xmath190 and @xmath501 with @xmath365 , @xmath498 replaced by @xmath491 , @xmath190 , respectively , in ( [ resolventexpansion ] ) and ( [ resolventexpansion2 ] ) .
note that @xmath503 with @xmath207-high probability by the bounded support condition .
then by ( [ boundh ] ) and ( [ resolventexpansion2 ] ) we see that @xmath504 with @xmath207-high probability . combining with ( [ mpbounds ] ) and ( [ diagonal ] ) , it is easy to check that there exist uniform constants @xmath505 such that with @xmath207-high probability , @xmath506 where we use that @xmath507 is uniformly bounded on @xmath107 . on the other hand , @xmath498 , @xmath501 and @xmath190
have the following trivial deterministic bound @xmath508 in this section , we use @xmath70 to denote the @xmath71-norm for any vector @xmath509 .
also we fix @xmath386 and @xmath510 throughout the following discussions .
the expressions will depend on @xmath386 , but we drop the subscripts for convenience .
the following lemma [ resolvent_error ] gives a trivial estimate for the remainder terms in ( [ resolventexpansion ] ) and ( [ resolventexpansion2 ] ) . by the definition of @xmath365 , for example
, we have @xmath513^m s)_{ab}= \sum_{(a_i , b_i ) \in \{(i,\mu ) , ( \mu , i)\}:1 \leq i \leq m } x_{i\mu}^m r_{a a_1 } r_{b_1 a_2 } \cdots s_{b_m b } .\ ] ] since there are only finitely many terms in the above sum for fixed @xmath514 , the conclusion follows immediately from ( [ 5bound2 ] ) and ( [ 5bound1 ] ) . [ def_operator1 ] for any two @xmath515 matrices @xmath65 and @xmath516 , we define @xmath517 as @xmath518 where @xmath488 is such that @xmath510 . in other words , we have @xmath519 when @xmath386 is fixed , we often drop the subscript @xmath386 and write @xmath520 for simplicity . also we denote the @xmath514-th power of @xmath65 under the @xmath521-product by @xmath522 , i.e. @xmath523 [ @xmath524 and @xmath525 notations ] [ def_operator2 ] for @xmath526 and @xmath527 , @xmath528 , we define @xmath529 where @xmath530 if @xmath531 and @xmath532 are products of matrix entries as above , then we define @xmath533 note that @xmath534 and @xmath535 are not linear operators acting on matrices , but just notations we use for simplification .
similarly , we define the following @xmath536 operator for the product of the entries of @xmath537 by @xmath538 where @xmath539 again , we will often drop the subscript @xmath386 whenever there is no confusion about @xmath386 . for @xmath540 and @xmath541 ,
it is easy to verify that @xmath542 where @xmath543 . for the second equality , note that @xmath544 is a sum of products of the entries of @xmath141 , where each product contains @xmath545 matrix entries
now we perform the green function comparison strategy .
the basic idea is to expand @xmath498 and @xmath190 in terms of @xmath501 using the resolvent expansions ( [ resolventexpansion ] ) and ( [ resolventexpansion2 ] ) , and compare the two expressions . we expect that the main terms will cancel since @xmath549 and @xmath550 have the same first four moments , and the error terms are small since @xmath549 and @xmath550 have support bounded by @xmath47 .
the key is the following lemma [ greenfunctionrepresent ] , whose proof can be found in ( * ? ? ?
* section 6 ) .
we denote @xmath551 [ green function representation theorem ] [ greenfunctionrepresent ] let @xmath552 as in ( [ 5bound1 ] ) and @xmath510 .
fix @xmath553 and @xmath554 .
then for any @xmath555 , we have @xmath556+\sum_{5 \leq \vert \mathbf{k } \vert \leq { 2 \zeta}/{\phi } , \mathbf k\in \mathbb z^s_+ } \mathcal{a } _ { \mathbf{k } } \mathbb{e } \ , \mathcal{p}_{\gamma,\mathbf{k } } \prod_{t=1}^s s_{a_t , b_t}+o(n^{-\zeta } ) , \label{onlyprove}\ ] ] where @xmath557 , @xmath558 , depend only on @xmath501 , and @xmath559 s depend both on @xmath501 and @xmath498 but are independent of @xmath560 , @xmath561 .
moreover , we have the estimate @xmath562 similarly , we have @xmath563+\sum_{5 \leq \vert \mathbf{k } \vert \leq { 2 \zeta}/{\phi } , \mathbf k \in \mathbb z_+^s } \mathcal{a } _ { \mathbf{k } } \mathbb{e } \ , \mathcal{\tilde{p}}_{\gamma,\mathbf{k}}\prod_{t=1}^s s_{a_t , b_t}+o(n^{-\zeta } ) , \label{onlyprove2}\ ] ] where @xmath564 , @xmath565 , again depend only on @xmath501 .
furthermore , as ( [ onlyprove ] ) , we also have @xmath566 where @xmath567 are independent of @xmath560 , @xmath561 , and @xmath568 it is clear that a result similar to lemma [ greenfunctionrepresent ] also holds for the product of @xmath190 s .
thus as in ( [ onlyprove ] ) , we define the notation @xmath569 , @xmath570 as follows : @xmath571+\sum_{5 \leq \vert \mathbf{k } \vert \leq { 2 \zeta}/{\phi } , \mathbf k\in \mathbb z^s_+ } \mathcal{a } _ { \mathbf{k}}^{\gamma , 0}\mathbb{e } \ , \mathcal{p}_{\gamma,\mathbf{k } } \prod_{t=1}^s s_{a_t , b_t}+o(n^{-\zeta}),\ ] ] @xmath572+\sum_{5 \leq \vert \mathbf{k } \vert \leq { 2 \zeta}/{\phi } , \mathbf k\in \mathbb z^s_+ } \mathcal{a } _ { \mathbf{k}}^{\gamma , 1}\mathbb{e } \ , \mathcal{p}_{\gamma,\mathbf{k } } \prod_{t=1}^s t_{a_t , b_t}+o(n^{-\zeta}).\ ] ] since @xmath573 , @xmath574 depend only on @xmath501 and @xmath575 , @xmath576 have the same first four moments , we get from ( [ greenpowers ] ) and ( [ greenpowert ] ) that @xmath577 then we obtain that @xmath578 by ( [ second ] ) and ( [ indexbound ] ) , the second term in ( [ teles_inequality ] ) is bounded by @xmath579 where we used @xmath580 and @xmath581 .
recall that @xmath582 is also a sum of products of @xmath141 . applying ( [ telescoping_2 ] ) again to the term @xmath583 and replacing @xmath584 in ( [ teles_inequality ] ) with @xmath585
, we obtain that @xmath586 together with ( [ teles_inequality ] ) , we have @xmath587 again using ( [ second ] ) and ( [ indexbound ] ) , it is easy to obtain that @xmath588 where we used that @xmath589 . repeating the above process ( [ teles_inequality])-([process_end ] ) @xmath590 times
, we obtain that @xmath591 where @xmath592 again using ( [ second ] ) , ( [ indexbound ] ) and @xmath593 , we get @xmath594 for some constant @xmath595 depending on @xmath596 .
recall that @xmath597 , @xmath598 and @xmath257 is a bounded support matrix with @xmath599 .
then by ( [ diagonal ] ) , with @xmath207-high probability , we have @xmath600 we now apply ( [ teles_inequality3 ] ) on @xmath601 with @xmath602 and @xmath603 .
note that @xmath604 is a finite sum of the products of the matrix entries of @xmath605 and @xmath606 , and moreover there are at least two off diagonal terms in each product ( see ( [ graphical ] ) ) .
thus let @xmath607 , we have @xmath608 which gives that @xmath609 for some constant @xmath610 .
plug it into ( [ teles_inequality3 ] ) , we conclude theorem [ thm_largebound ] . for simplicity , instead of ( [ keyeyeyey ] ) , we shall prove that @xmath611 the proof for ( [ keyeyeyey ] ) is exactly the same but with slightly heavier notations ( in the product of @xmath612 terms , half of them are normal and the other half are complex conjugates ) .
we start with ( [ teles_inequality3 ] ) .
define a function @xmath613 with @xmath614 since @xmath615 and @xmath616 are independent of @xmath617 and @xmath618 ( @xmath619 ) , we may consider a linear combination of ( [ teles_inequality3 ] ) with coefficients given by @xmath613 .
moreover with ( [ onlyprove2 ] ) , we can extend ( [ teles_inequality3 ] ) to the product of @xmath537 terms for some constant @xmath620 , i.e. @xmath621 if we take @xmath622 and @xmath623 , it is easy to check that @xmath624 now to conclude ( [ simpler_pf ] ) , it suffices to control the second term on the right hand side of ( [ key22 ] ) .
we consider the terms for @xmath625 satisfying ( [ indexassumption ] ) . by definition of @xmath626 , ( [ average_comparison1 ] )
is a sum of at most @xmath627 products of @xmath628 and @xmath629 terms , where the total number of @xmath628 and @xmath629 in each product is @xmath630 . due to
the rough bound ( [ 5boundt ] ) , ( [ average_comparison1 ] ) is always bounded by @xmath631 . then with the assumption that ( [ keybounds ] ) and ( [ 5bound1 ] ) hold with @xmath207-high probability with @xmath225 , we see that the event that ( [ keybounds ] ) or ( [ 5bound1 ] ) does not hold is negligible .
furthermore , for each product in ( [ average_comparison1 ] ) and any @xmath632 , there are two @xmath633 s in the indices of @xmath141
. these two @xmath633 s can only appear as ( 1 ) @xmath634 in the product , or ( 2 ) @xmath635 , where @xmath636 come from some @xmath637 and @xmath638 ( @xmath639 ) via @xmath536 ( see definition [ def_operator2 ] ) .
then after averaging over @xmath640 @xmath641 this term becomes ( 1 ) @xmath642 , which is bounded by @xmath643 by ( [ keybounds ] ) , or ( 2 ) @xmath644 , which is bounded by @xmath645 by ( [ keybounds ] ) .
for all other @xmath141 factors in the product with no @xmath633 s , we control them by @xmath51 using ( [ 5bound1 ] ) .
then for any fixed @xmath646 , @xmath625 , we have proved that @xmath647 together with ( [ key22 ] ) , we conclude ( [ simpler_pf ] ) . for simplicity ,
we only prove ( [ bdbd ] ) .
the proof for ( [ bdbd1 ] ) is similar . by ( [ eq_gsq1 ] )
, we have @xmath648 hence , it is equivalent to show that @xmath649 for @xmath650 with @xmath651 and @xmath652 . as in ( [ r ] ) , we denote @xmath653 by cauchy interlacing theorem , it is easy to check that @xmath654 together with ( [ mpbounds ] ) and ( [ newmpbounds ] ) , we get with high probability @xmath655 by ( [ diagonal ] ) , ( [ 5bound2 ] ) and the expansion ( [ resolventexpansion2 ] ) , we get that with high probability , @xmath656 moreover , by ( [ 5boundt ] ) we have the trivial bounds @xmath657 since the bad events hold with exponentially small probability , we simply ignore the bad events in our proof . applying the lindeberg replacement strategy again , we have @xmath658.\ ] ] from the taylor expansion , we have @xmath659 where @xmath660 lies between @xmath661 and @xmath662 .
let @xmath510 .
we perform the expansion ( [ resolventexpansion ] ) and use ( [ graphical ] ) to get that @xmath663 using this expansion and bound ( [ 5bound1 ] ) , we have that with @xmath207-high probability , @xmath664 where @xmath665 from the above definition , we have the rough bound @xmath666 by lemma [ propsotion_poperator ] and ( [ bbbbbbbbbbb ] ) , the @xmath667 terms in ( [ bigsum ] ) can be bounded by @xmath668 with @xmath207-high probability .
hence with @xmath207-high probability , @xmath669 similarly , we also have @xmath670 using ( [ reduced1 ] ) and choose @xmath671 , we obtain @xmath672 and @xmath673 with high probability .
to control the second term in ( [ rsequation ] ) , we need the following lemma . given ( [ lem_skip ] ) , with markov inequality we find that for any fixed @xmath674 , @xmath675 , @xmath679 holds with probability with @xmath680 for any @xmath205 , where we used that @xmath681 . by cauchy - schwarz inequality
, we have ( recall ( [ conditiona3 ] ) ) @xmath682 combining ( [ rsequation ] ) , ( [ productbound ] ) and ( [ holderbound ] ) , we can find a constant @xmath58 such that @xmath683 for sufficiently large @xmath11 independent of @xmath386 . since @xmath660 is between @xmath661 and @xmath662 , we have @xmath684 with high probability by ( [ 578 ] ) . together with ( [ taylorbound_1 ] ) and the assumption ( [ fcondtion ] )
, we get @xmath685 \right\vert \leq n^{-\phi+c_4\epsilon},\ ] ] for some @xmath445 .
we have a similar estimate for @xmath686 $ ] . from ( [ taylorexpansion ] ) and ( [ taylorexp1 ] )
, we see that it remains to deal with the first term in ( [ taylorexpansion ] ) . using ( [ rsequation ] ) , ( [ rsequation2 ] ) and the fact that the first four moments of @xmath549 and @xmath687 match , we obtain that @xmath688-\mathbb{e}\left[f^{(s)}(x^r)(x^t - x^r)^s \right ] \right \vert \nonumber \\ & \leq \left| \sum_{k=5}^{{9}/{\phi } } \
\sum_{\sum_{t=1}^s \vert \mathbf{k}_t \vert = k } \sum_{\mathbf{k}_t \in i_{{3}/{\phi},k}^{2s}}\mathbb{e } \prod_{t=1}^s \left(\mathcal{p}_{\gamma , \mathbf{k}_t } x^r\right)\right| \left(\left\vert \mathbb{e}(-x_{i \mu})^{k } \right\vert+ \left\vert \mathbb{e}(\tilde{x}_{i \mu})^{k } \right\vert \right)+o(cn^{-3}).\end{aligned}\ ] ] recall that ( [ holderbound ] ) and ( [ conditiona3 ] ) hold for @xmath689 and @xmath550 , @xmath549 has support bounded by @xmath47 , and @xmath550 has support bounded by @xmath690 .
then it is easy to check that @xmath691 and @xmath692 for @xmath693 . using ( [ productbound ] )
, we obtain that @xmath694-\mathbb{e}\left[f^{(s)}(x^r)(x^t - x^r)^s\right ] \right\vert \leq n^{-2-\phi+c_4\epsilon } , \
1 \leq s \leq 2\ ] ] with this estimate , it is easy to conclude the proof .
a major part of this appendix is devoted to prove the entrywise local law ( [ diagonal ] ) and the averaged local law ( [ mpbounds ] ) .
the other results of lemma [ lem_small ] are mostly consequences of these two local laws , and we will describe briefly their proof at the end of this appendix . in @xcite ,
the authors have proved slightly weaker forms of the entrywise and averaged local laws for @xmath141 under weaker assumptions ( in fact , our @xmath141 is a simple special case of the more general green functions considered in that paper ) .
we will basically follow the approach in @xcite , but change some key estimates with our stronger assumptions and some tools in @xcite . for @xmath695 , we define the minor @xmath696 obtained by removing all rows and columns of @xmath79 indexed by @xmath697 .
note that we keep the names of indices of @xmath141 when defining @xmath698 , i.e. @xmath699 .
correspondingly , we define the green function @xmath700 and the partial traces @xmath701 we will abbreviate @xmath702 , @xmath703 , and @xmath704 * for @xmath705 and @xmath706 , we have @xmath707 * for @xmath708 and @xmath709 , we have @xmath710 for @xmath705 and @xmath706 , we have @xmath711 * for @xmath712 and @xmath713 , @xmath714 * all of the above identities hold for @xmath715 instead of @xmath141 for @xmath716 .
[ ward_id ] fix any @xmath717 .
the following estimates hold uniformly for any @xmath552 .
we have @xmath718 and the following identities @xmath719 all of the above estimates remain true for @xmath715 instead of @xmath141 for any @xmath695 . [ lemma_im ]
these estimates and identities can be proved through simple calculations using ( [ green2 ] ) , ( [ spectral1 ] ) and ( [ spectral2 ] ) .
we refer the reader to , for example , ( * ? ? ?
* lemma 4.6 ) and ( * ? ? ?
* lemma 3.5 ) . for @xmath722 , we have @xmath723 ) , in the second and third steps the equality ( [ eq_gsq1 ] ) .
similarly , using ( [ resolvent8 ] ) and ( [ eq_gsq3 ] ) we get @xmath724 is uniformly bounded on @xmath107 and @xmath725 , we can prove ( [ m_t ] ) by induction on the indices in @xmath726 .
the proof for @xmath727 is similar .
the following large deviation bound for bounded supported random variables is lemma 3.8 of @xcite .
it constitutes the main difference between our proof and the one in @xcite , where the authors used a large deviation bound for random variables with arbitrarily high moments .
[ largederivation ] let @xmath728 , @xmath729 be independent families of centered and independent random variables , and @xmath730 , @xmath731 be families of deterministic complex numbers .
suppose the entries @xmath732 and @xmath733 have variance @xmath734 and satisfies the bounded support condition ( [ eq_support ] ) with @xmath735 for some @xmath131 .
then for any fixed @xmath119 , the followings hold with @xmath118-high probability : @xmath736 , \\
\left\vert \sum_{i , j } x_i b_{ij } y_j \right\vert & \leq \varphi^{2\xi}\left[q^2 b_d + qb_o + \frac{1}{n}\left(\sum_{i\ne j } |b_{ij}|^2\right)^{{1}/{2 } } \right ] , \\ \left\vert \sum_{i } \bar x_i b_{ii } x_i - \sum_{i } ( \mathbb e|x_i|^2 ) b_{ii } \right\vert & \leq \varphi^{\xi}q b_d , \ \
\left\vert \sum_{i\ne j } \bar x_i b_{ij } x_j \right\vert \leq \varphi^{2\xi}\left[qb_o + \frac{1}{n}\left(\sum_{i\ne j } |b_{ij}|^2\right)^{{1}/{2 } } \right],\end{aligned}\ ] ] where @xmath737 for the proof in this section , we fix @xmath225 .
our goal is to prove that @xmath141 is close to @xmath199 in the sense of entrywise and averaged local laws .
hence it is convenient to introduce the following random control parameters . in analogy to ( * ? ? ?
* section 3 ) and ( * ? ? ?
* section 5 ) , we introduce the @xmath743 variables @xmath744 where @xmath745:=\mathbb e[\cdot\mid h^{(a)}],$ ] i.e. it is the partial expectation in the randomness of the @xmath403-th row and column of @xmath79 . by ( [ resolvent2 ] ) , we have @xmath746 and @xmath747 the following estimate plays a key role in the proof of local laws . [ z_lemma ] fix @xmath717 and @xmath748 , and define the @xmath405-dependent event @xmath749 .
then there exists @xmath58 such that the following estimates hold for all @xmath750 and @xmath552 with @xmath118-high probability : @xmath751 and @xmath752 apply the large deviation lemma [ largederivation ] to @xmath753 in ( [ zi ] ) , we get that on @xmath754 , @xmath755 \nonumber\\ = & c \varphi^{2\xi } \left [ q + \frac{1}{n}\left ( { \sum_\mu \frac{{\mathop{\rm im}\nolimits}\ , g_{\mu\mu}^{(i ) } } { \eta } } \right)^{1/2}\right ] = c \varphi^{2\xi } \left [ q + \sqrt { \frac { { { \rm{im}}}\ , m_2^{(i ) } } { n\eta } } \right ] .\end{aligned}\ ] ] holds with @xmath118-high probability , where we used ( [ eq_gsq1 ] ) and the fact that @xmath756 on event @xmath754 . now using the bound ( [ m_t ] ) and the definitions ( [ eqn_randomerror ] ) , ( [ eq_defpsitheta ] ) , we get that @xmath757 together with the estimate ( [ estimate_zi ] ) , we conclude that @xmath758 with @xmath118-high probability .
similarly , we can prove the same estimate for @xmath759 .
if @xmath760 , we always have @xmath756 by ( [ eq_gbound ] ) .
then the above proof immediately gives that @xmath761 with @xmath118-high probability .
similar as above , using ( [ resolvent3 ] ) and lemmas [ ward_id]-[largederivation ] , we can prove that @xmath762 holds with @xmath118-high probability .
it remains to prove the bound for @xmath763 and @xmath764 .
using ( [ resolvent6 ] ) , the bounded support condition ( [ eq_support ] ) for @xmath765 , the bound @xmath756 on @xmath754 , and lemma [ largederivation ] , we get that with @xmath118-high probability , @xmath766 \nonumber\\ & \le c\varphi^{2\xi}\left[q+ \frac{1}{n}\left ( \sum_{\nu}\left({g}^{(i\mu)}_{\nu \nu } + \frac{\bar z}{\eta } { { \rm{im}}}\ , g_{\nu\nu}^{(i\mu)}\right)\right)^{1/2 } \right ] \le c\varphi^{2\xi}\left[q+\sqrt { \frac{|m_2^{(i\mu)}|}{n } } + \sqrt{\frac { { { \rm{im}}}\ , m_2^{(i\mu)}}{n\eta } } \right ] , \label{off_larged}\end{aligned}\ ] ] where in the third step we used ( [ eq_gsq3 ] ) . as in ( [ m2psi ] ) , we can show that @xmath767 for the other term , we have @xmath768 where we used ( [ m_t ] ) , and that @xmath769 since @xmath770 and @xmath771 by lemma [ lem_mbehavior ] .
plug the two estimates ( [ estimatel1 ] ) and ( [ estimatel2 ] ) into ( [ off_larged ] ) , we conclude that @xmath772 with @xmath118-high probability . combining with ( [ 2blocks ] ) , we get the estimate in ( [ zestimate1 ] ) for @xmath773 .
finally , the estimate ( [ zestimate2 ] ) can be proved in a similar way with the bound @xmath774 . as in @xcite , our proof of the local laws
start with an analysis of the self - consistent equations .
note that @xmath775 and @xmath776 in the next lemma correspond to the equations ( [ selfm1c ] ) and ( [ moself ] ) , respectively .
[ lemm_selfcons_weak ] fix @xmath225 and @xmath777
. then there exists @xmath58 such that the followings hold uniformly in @xmath778 with @xmath207-high probability : @xmath779 and @xmath780 where @xmath754 is defined in lemma [ z_lemma ] and @xmath781 moreover , we have finer estimates @xmath782 + o\left(\varphi^{4\xi_1}(q^2+\psi^2_\theta ) \right ) , \ \ { \mathbf 1}(\xi)\mathcal d_2(z , m_2 ) = \mathbf 1(\xi)[z_2]+ o\left(\varphi^{4\xi_1 } ( q^2+\psi^2_\theta)\right),\ \label{selfcons_improved}\ ] ] with @xmath207-high probability , where @xmath783:=\frac{1}{m}\sum_{i\in \mathcal i_1 } z_i , \ \ [ z_2]:=\frac{1}{n}\sum_{\mu \in \mathcal i_2 } z_\mu.\ ] ] by ( [ resolvent2 ] ) and the definition of @xmath784 , we have @xmath785 where @xmath786 by ( [ m_t ] ) and lemma [ z_lemma ] , we have with @xmath207-high probability @xmath787 then using ( [ self_gii ] ) , we get that @xmath788 with @xmath207-high probability .
this implies that @xmath789 with @xmath207-high probability . plugging this estimate into ( [ self_gii ] ) and using that @xmath790 on @xmath754 yields @xmath791 with @xmath207-high probability .
together with the relation ( [ barm ] ) , the above equation gives the estimate for @xmath775 in ( [ selfcons_lemm ] ) .
the proof for @xmath776 is similar .
to prove ( [ selfcons_improved ] ) , we write @xmath792 after taking the average @xmath793 , the second term on the right - hand side vanishes and the third term provides a @xmath794 factor by ( [ average_bound ] ) . on the other hand , using ( [ resolvent8 ] ) and ( [ zestimate1 ] ) we get @xmath795 with @xmath207-high probability . thus taking average of ( [ self_gii ] ) yields @xmath796= \mathbf 1(\xi)\left [ - 1 - m_2 + [ z_1 ] + o\left(\varphi^{4\xi_1 } ( q + \psi_\theta)^2 \right)\right],\ ] ] with @xmath207-high probability .
again plug into the relation ( [ barm ] ) , we obtain the estimate for @xmath775 in ( [ selfcons_improved ] ) .
the proof for @xmath776 is similar . finally , we prove ( [ selfcons_lemm2 ] ) . using the bound @xmath797 ,
it is easy to see that @xmath798 . combining with ( [ immc ] )
, we immediately get that @xmath799 and hence @xmath800 with @xmath207-high probability . from ( [ self_gii ] ) and ( [ barm ] ) we get that @xmath801
it thus follows that @xmath802 ) , we have @xmath803 which gives @xmath804 since @xmath805 for all @xmath232 .
thus we have @xmath806 with @xmath207-high probability . with this estimate and ( [ epsilonl ] ) , we get from ( [ self_m1r ] ) that @xmath807 with @xmath207-high probability .
together with ( [ estimate_m1l ] ) , the above equation gives the estimate for @xmath775 in ( [ selfcons_lemm2 ] ) .
the proof for @xmath776 is similar . to get the desired local laws from lemma [ lemm_selfcons_weak ] , we need the following lemma [ stability ] , which gives the stability of @xmath775 and @xmath776 .
roughly , it states that if @xmath808 is small , then @xmath809 is close to @xmath810 . for an arbitrary @xmath811 , we define the discrete set @xmath812\cap ( n^{-10}\mathbb n)\ } .\end{aligned}\ ] ] thus , if @xmath813 then @xmath814 , and if @xmath815 then @xmath816 is a 1-dimensional lattice with spacing @xmath817 plus the point @xmath405 .
obviously , we have @xmath818 .
the following lemma is stated as definition 5.4 of @xcite and lemma 4.5 of @xcite .
[ stability ] the self - consistent equation @xmath775 in ( [ def_d12 ] ) is stable on @xmath819 in the following sense .
suppose that @xmath820 for @xmath811 and that @xmath53 is lipschitz continuous with lipschitz constant @xmath821 .
suppose moreover that for each fixed @xmath349 , the function @xmath822 is non - increasing for @xmath422 .
suppose that @xmath823 is the stieltjes transform of a probability measure .
let @xmath811 and suppose that for all @xmath824 we have @xmath825 then we have @xmath826 for some constant @xmath58 independent of @xmath405 and @xmath11 , where @xmath827 is defined in ( [ kappa ] ) .
similarly , the self - consistent equation @xmath776 in ( [ def_d12 ] ) is also stable on @xmath819 .
note that by this lemma and ( [ selfcons_lemm2 ] ) , we immediately get that @xmath828 with @xmath207-high probability . then using ( [ zestimate2 ] )
, we obtain that @xmath829 with @xmath207-high probability .
also as in ( [ average_bound ] ) , we can prove that @xmath830 with @xmath207-high probability , which gives the diagonal estimate together with ( [ average_l ] ) . comparing with lemma [ lem_small ]
, we see that these bound are almost optimal for @xmath760 .
now it remains to deal with the @xmath831 case ( in particular , the @xmath832 case ) .
one can first prove the following weaker result .
one can prove this lemma using a continuity argument as in ( * ? ? ?
* section 4.1 ) or ( * ? ? ?
* section 3.6 ) .
the key inputs for the proof are lemmas [ z_lemma]-[stability ] , diagonal estimate ( [ average_bound ] ) and the estimates ( [ average_l])-([diag_l ] ) in the @xmath834 case . all the other parts of the proof are essentially the same .
to get the strong local laws given in lemma [ lem_small ] , we need better bounds on @xmath835 $ ] and @xmath836 $ ] in ( [ selfcons_improved ] ) .
they follow from the following abstract decoupling lemma ( or the fluctuation averaging lemma " ) .
it is stated as lemma 4.1 of @xcite , and we do not repeat the proof here . [ abstractdecoupling ] fix @xmath119 , and suppose @xmath837 and that there exists @xmath838 with @xmath839 such that we have with @xmath118-high probability @xmath840 where @xmath386 is a deterministic function satisfying @xmath841 .
let @xmath842 .
then we have with @xmath843-high probability , @xmath844(z)|+ |[z_2](z)| \le \varphi^{18\xi } \left(q^2 + \frac{1}{(n\eta)^2 } + \frac{{{\rm{im}}}(m_{1c}(z)+m_{2c}(z))+ \gamma(z)}{n\eta } \right),\ ] ] for @xmath845 . during the proof ,
we fix @xmath846 and take @xmath847 hence we have @xmath848 and @xmath849 . then to prove ( [ diagonal ] )
, it suffices to prove @xmath850 with @xmath118-high probability . for notational convenience
, we shall denote @xmath851 . by lemmas [ alem_weak ]
, we have that @xmath754 holds with @xmath852-high probaility . then together with lemma [ abstractdecoupling ] and ( [ selfcons_improved ] )
, we get that with @xmath853-high probability , @xmath854 \\ & \le c\left[\varphi^{20\tilde \xi } \left(q^2 + \frac{1}{(n\eta)^{4/3}}\right ) + \varphi^{18\tilde \xi } \frac{{{\rm{im}}}\ , m_c}{n\eta } \right ] , \end{aligned}\ ] ] where we used young s inequality for the @xmath855 term . now applying lemma [ stability ] to the above estimate
, we get that with @xmath853-high probability , @xmath856 where we used ( [ squarerootbehavior ] ) in the second step .
we can prove a similar estimate for @xmath857 , so we have @xmath858 then using lemma [ z_lemma ] and ( [ average_bound ] ) , it is easy to obtain that @xmath859 uniformly in @xmath860 with @xmath861-high probability , which is a better bound than the one in ( [ localweakm ] ) .
we can repeat this process @xmath10 times , each iteration yields a stronger bound on @xmath862 which holds with a smaller probability . more specifically ,
suppose that after @xmath232 iterations we get the bound @xmath863 uniformly in @xmath860 with @xmath864-high probability .
then by lemma [ abstractdecoupling ] and ( [ selfcons_improved ] ) , we have with @xmath865-high probability , @xmath866 \\ & \le c\left[\varphi^{38\tilde \xi } \left(q^2 + \frac{1}{(n\eta)^{2-\tau}}\right ) + \varphi^{18\tilde \xi } \frac{{{\rm{im}}}\ , m_c}{n\eta } \right ] .\end{aligned}\ ] ] applying lemma [ stability ] , we get that with @xmath865-high probability , @xmath867 with the similar estimate for @xmath857 , we get @xmath868 again using lemma [ z_lemma ] and ( [ average_bound ] ) , we obtain that @xmath869 uniformly in @xmath860 with @xmath870-high probability . comparing with ( [ iteration1 ] )
, we see that the power of @xmath871 is increased from @xmath872 to @xmath873 , and moreover , there is no constant @xmath51 appearing in ( [ iteration2 ] ) .
thus after @xmath10 iterations , we get @xmath874 uniformly in @xmath860 with @xmath875-high probability . taking @xmath876
such that @xmath877 we can conclude ( [ goal_law1 ] ) and hence ( [ diagonal ] ) .
finally to get ( [ mpbounds ] ) , we plug ( [ goal_law1 ] ) into lemma [ abstractdecoupling ] and use lemma [ stability ] .
choose @xmath878 with @xmath879 . by ( [ diagonal ] ) , we have @xmath880 then using the spectral decomposition ( [ spectral1 ] ) , we get @xmath881 choosing @xmath882 in ( [ spectraldecomp ] ) yields that , with @xmath207-high probability , @xmath883 for all @xmath232 .
the proof for @xmath884 is similar .
* acknowledgements . *
the authors would like to thank professor jeremy quastel and professor jun yin for fruitful discussions and valuable suggestions , which have significantly improved the paper .
the first author also want to thank professor jun yin for the hospitality when he visited madison .
j. w. silverstein .
the stieltjes transform and its role in eigenvalue behavior of large dimensional random matrices .
random matrix theory and its applications , lecture notes series .
world scientific , singapore , 2009 . | in this paper , we prove a necessary and sufficient condition for the edge universality of covariance matrices , which is conjectured by pillai and yin in @xcite .
consider the sample covariance matrices of the form @xmath0 , where @xmath1 is an @xmath2 rectangular matrix with i.i.d entries @xmath3 satisfying @xmath4 and @xmath5 . under the assumption @xmath6 , we prove that the tracy - widom law holds at the soft edge ( i.e. , for the largest eigenvalues ) if and only if @xmath7 .
this condition was first proposed for wigner matrices by lee and yin @xcite . |
for many years now , spin physics has played a very prominent role in qcd .
the field has been carried by the hugely successful experimental program of polarized deeply - inelastic lepton - nucleon scattering ( dis ) , and by a simultaneous tremendous progress in theory .
this talk summarizes some of the interesting new developments in spin physics in the past roughly two years .
as we will see , there have yet again been exciting new data from polarized lepton - nucleon scattering , but also from the world s first polarized @xmath0 collider , rhic .
there have been very significant advances in theory as well .
it will not be possible to cover all developments .
i will select those topics that may be of particular interest to the attendees of a conference in the `` dis '' series .
until a few years ago , polarized inclusive dis played the dominant role in qcd spin physics @xcite . at the center of attention was the nucleon s spin structure function @xmath1 .
[ fig1 ] shows a recent compilation @xcite of the world data on @xmath1 .
these data have provided much interesting information about the nucleon and qcd .
for example , they have given direct access to the helicity - dependent parton distribution functions of the nucleon , @xmath2 polarized dis actually measures the combinations @xmath3 . from @xmath4 extrapolation of the structure functions for proton and neutron targets it has been possible to test and confirm the bjorken sum rule @xcite .
polarized dis data , when combined with input from hadronic @xmath5 decays , have allowed to extract the unexpectedly small
nucleon s axial charge @xmath6 , which to lowest order unambiguously coincides with the quark spin contribution to the nucleon spin @xcite . the results from polarized inclusive dis have also led us to identify the next important goals in our quest for understanding the spin structure of the nucleon .
the measurement of gluon polarization @xmath7 rightly is a main emphasis at several experiments in spin physics today , since @xmath8 could be a major contributor to the nucleon spin .
also , more detailed understanding of polarized quark distributions is clearly needed ; for example , we would like to know about flavor symmetry breakings in the polarized nucleon sea , details about strange quark polarization , and also about the small-@xmath9 and large-@xmath9 behavior of the densities .
again , these questions are being addressed by current experiments .
finally , we would like to find out how much orbital angular momentum quarks and gluons contribute to the nucleon spin .
ji showed @xcite that their total angular momenta may be extracted from deeply - virtual compton scattering , which has sparked much experimental activity also in this area .
there are several lepton - nucleon scattering fixed - target experiments around the world with dedicated spin physics programs
. this will not be a complete list ; i will mention only those that play a role in this talk .
hermes at desy uses hera s 27.5 gev polarized electron beam on polarized targets .
they have just completed a run with a transversely polarized target .
semi - inclusive dis ( sidis ) measurements are one particular strength of hermes .
compass at cern uses a 160 gev polarized muon beam .
their main emphasis is measuring gluon polarization ; they have completed their first run .
there is also a very large spin program at jefferson lab , involving several experiments .
large-@xmath9 structure functions and the dvcs reaction are just two of many objectives there . finally , an experiment e161 at slac aims at measuring @xmath8 in photoproduction , but has unfortunately been put on hold awaiting funding . for the more distant future , there are plans to develop a polarized electron - proton _ collider _ at bnl , erhic @xcite . a new milestone has been reached in spin physics by the advent of the first polarized proton - proton collider , rhic at bnl . by now , two physics runs with polarized protons colliding at @xmath10 gev have been completed , and exciting first results are emerging .
we will see one example toward the end of this talk .
all components crucial for the initial phase of the spin program with beam polarization up to 50% are in place @xcite .
this is true for the accelerator ( polarized source , siberian snakes , polarimetry by proton - carbon elastic scattering ) as well as for the detectors .
rhic brings to collision 55 bunches with a polarization pattern , for example , @xmath11 in one ring and @xmath12 in the other , which amounts to collisions with different spin combinations every 106 nsec .
it has been possible to maintain polarization for about 10 hours
. there is still need for improvements in polarization and luminosity for future runs .
the two larger rhic experiments , phenix and star , have dedicated spin programs focusing on precise measurements of @xmath8 , quark polarizations by flavor , phenomena with transverse spin , and many others . as mentioned above , the measurement of @xmath8 is a main goal of several experiments .
the gluon density affects the @xmath13-evolution of the structure function @xmath1 , but the limited lever arm in @xmath13 available so far has left @xmath8 virtually unconstrained .
one way to access @xmath8 in lepton - nucleon scattering is therefore to look at a less inclusive final state that is particularly sensitive to gluons in the initial state .
one channel , to be investigated by compass in particular , is heavy - flavor production via the photon - gluon fusion process @xcite .
an alternative reaction is @xmath14 , where the two hadrons in the final state have large transverse momentum @xcite .
rhic will likely dominate the measurements of @xmath8 .
several different processes will be investigated @xcite that are sensitive to gluon polarization : high-@xmath15 prompt photons @xmath16 , jet or hadron production @xmath17 , @xmath18 , and heavy - flavor production @xmath19 . in addition , besides the current @xmath10 gev , also @xmath20 gev will be available at a later stage .
all this will allow to determine @xmath21 in various regions of @xmath9 , and at different scales .
one can compare the @xmath8 extracted in the various channels , and hence check its universality implied by factorization theorems . in this way
, we will also likely learn a lot more about high-@xmath15 reactions in qcd .
we emphasize that for all the reactions relevant at rhic we now know the next - to - leading order ( nlo ) qcd corrections to the underlying hard scatterings of polarized partons @xcite .
this significantly improves the theoretical framework , since it is known from experience with the unpolarized case that the corrections are indispensable in order to arrive at quantitative predictions for hadronic cross sections .
for instance , the dependence on factorization and renormalization scales in the calculation is much reduced when going to nlo . therefore ,
only with knowledge of the nlo corrections will one be able to extract @xmath8 reliably .
figure [ fig2 ] shows nlo predictions @xcite for the double - spin asymmetry @xmath22 for the reaction @xmath23 at rhic , using various different currently allowed parameterizations @xcite of @xmath21 .
it also shows the statistical errors bars expected for a measurement by phenix in @xmath23 with lower polarization and luminosity were reported by phenix @xcite .
] under the assumption of 50% beam polarizations and 7/pb integrated luminosity .
it is evident that the prospects for determining @xmath8 in this reaction , and in related ones , are excellent .
we stress that phenix has recently presented a measurement of the unpolarized high-@xmath15 @xmath24 cross section @xcite that agrees well with an nlo perturbative - qcd calculation over the whole range of @xmath15 accessed .
this provides confidence that the theoretical hard scattering framework used for fig .
[ fig2 ] is indeed adequate . as mentioned earlier , inclusive dis via photon exchange
only gives access to the combinations @xmath25 .
there are at least two ways to distinguish between quark and antiquark polarizations , and also to achieve a flavor separation .
semi - inclusive measurements in dis are one possibility , explored by smc @xcite and , more recently and with higher precision , by hermes @xcite .
one detects a hadron in the final state , so that instead of @xmath26 the polarized dis cross section becomes sensitive to @xmath27 , for a given quark flavor . here ,
the @xmath28 are fragmentation functions , with @xmath29 .
[ fig3 ] shows the latest results on the flavor separation by hermes @xcite , obtained from their lo monte - carlo code based `` purity '' analysis . within the still fairly large uncertainties , they are not inconsistent with the large negative polarization of @xmath30 in the sea that has been implemented in many determinations of polarized parton distributions from inclusive dis data @xcite ( see curves in fig . [ fig3 ] ) . on the other hand
, there is no evidence either for a large negative strange quark polarization . for the region @xmath31
, the extracted @xmath32 integrates @xcite to the value @xmath33 , while analyses of inclusive dis prefer an integral of about -0.025 .
there is much theory activity currently on sidis , focusing also on possible systematic improvements to the analysis method employed in @xcite , among them nlo corrections , target fragmentation , and higher twist contributions @xcite .
we note that at rhic @xcite one will use @xmath34 production to determine @xmath35 with good precision , making use of parity - violation .
comparisons of such data taken at much higher scales with those from sidis will be extremely interesting .
new interesting information on the polarized quark densities has also recently been obtained at high @xmath9 .
the hall a collaboration at jlab has published their data for the neutron asymmetry @xmath36 @xcite , shown in fig .
[ fig4 ] ( left ) .
the new data points show a clear trend for @xmath36 to turn positive at large @xmath9 .
such data are valuable because the valence region is a particularly useful testing ground for models of nucleon structure .
the right panel of fig .
[ fig4 ] shows the extracted polarization asymmetry for @xmath37 .
the data are consistent with constituent quark models @xcite predicting @xmath38 at large @xmath9 , while `` hadron helicity conservation '' predictions based on perturbative qcd and the neglect of quark orbital angular momentum @xcite give @xmath39 and tend to deviate from the data , unless the convergence to 1 sets in very late .
besides the unpolarized and the helicity - dependent densities , there is a third set of twist-2 parton distributions , transversity @xcite . in analogy with eq .
( [ eq1 ] ) they measure the net number ( parallel minus antiparallel ) of partons with transverse polarization in a transversely polarized nucleon : @xmath40 in a helicity basis , one finds @xcite that transversity corresponds to a helicity - flip structure , as shown in fig .
this precludes a gluon transversity distribution at leading twist .
it also makes transversity a probe of chiral symmetry breaking in qcd @xcite : perturbative - qcd interactions preserve chirality , and so the helicity flip required to make transversity non - zero must primarily come from soft non - perturbative interactions for which chiral symmetry is broken .
measurements of transversity are not straightforward .
again the fact that perturbative interactions in the standard model do not change chirality ( or , for massless quarks , helicity ) means that inclusive dis is not useful .
collins , however , showed @xcite that properties of fragmentation might be exploited to obtain a `` transversity polarimeter '' : a pion produced in fragmentation will have some transverse momentum with respect to the fragmenting parent quark .
there may then be a correlation of the form @xmath41 .
the fragmentation function associated with this correlation is the collins function .
the phase is required by time - reversal invariance .
the situation is depicted in fig .
the collins function would make a _ leading - power _
@xcite contribution to the single - spin asymmetry @xmath42 in the reaction @xmath43 : @xmath44 where @xmath45 ( @xmath46 ) is the angle between the lepton plane and the @xmath47 plane ( and the transverse target spin ) .
as is evident from eq .
( [ eq3 ] ) , this asymmetry would allow access to transversity if the collins functions are non - vanishing .
a few years ago , hermes measured the asymmetry for a longitudinally polarized target @xcite . for finite @xmath48 , the target spin then has a transverse component @xmath49 relative to the direction of the virtual photon , and the effect may still be there , even though it is now only one of several `` higher twist '' contributions @xcite .
if `` intrinsic '' transverse momentum in the fragmentation process plays a crucial role in the asymmetry for @xmath50 , a natural question is whether @xmath51 in the initial state can be relevant as well .
sivers suggested @xcite that the @xmath51 distribution of a quark in a transversely polarized hadron could have an azimuthal asymmetry , @xmath52 , as shown in fig .
[ fig7 ] .
there is a qualitative difference between the collins and sivers functions , however . while phases will always arise in strong interaction final - state fragmentation
, one does not expect them from initial ( stable ) hadrons , and the sivers function appears to be ruled out by time - reversal invariance of qcd @xcite . until recently
, it was therefore widely believed that origins of single - spin asymmetries as in @xmath53 and other reactions were more likely to be found in final - state fragmentation effects than in initial state parton distributions .
however , then came a model calculation @xcite that found a leading - power asymmetry in @xmath54 not associated with the collins effect .
it was subsequently realized @xcite that the calculation of @xcite could be regarded as a model for the sivers effect .
it turned out that the original time - reversal argument against the sivers function is invalidated by the presence of the wilson lines in the operators defining the parton density .
these are required by gauge invariance and had been neglected in @xcite . under time reversal , however , future - pointing wilson lines turn into past - pointing ones , which changes the time reversal properties of the sivers function and allows it to be non - vanishing . now ,
for a `` standard '' , @xmath51-integrated , parton density the gauge link contribution is unity in the @xmath55 gauge , so one may wonder how it can be relevant for the sivers function .
the point , however , is that for the case of @xmath51-dependent parton densities , a gauge link survives even in the light - cone gauge , in a transverse direction at light - cone component @xmath56 @xcite .
thus , time reversal indeed does not imply that the sivers function vanishes .
the same is true for a function describing transversity in an unpolarized hadron @xcite .
it is intriguing that these new results are based entirely on the wilson lines in qcd .
if the sivers function is non - vanishing , it will for example make a leading - power contribution to @xmath53 , of the form @xmath57 this is in competition with the collins function contribution , eq .
( [ eq3 ] ) ; however , the azimuthal angular dependence is discernibly different .
hermes has just completed a run with transverse polarization , and preliminary results are expected soon .
we note that the collins function may also be determined separately from an azimuthal asymmetry in @xmath58 annihilation @xcite .
it was pointed out @xcite that comparisons of dis and the drell - yan process will be particularly interesting : from the properties of the wilson lines it follows that the sivers functions relevant in dis and in the drell - yan process have opposite sign , violating universality of the distribution functions .
this is a striking prediction awaiting experimental testing . for work on the process ( in)dependence of the collins function ,
see @xcite ; recent model calculations of the function in the context of the gauge links may be found in @xcite .
originally , the sivers function was proposed @xcite as a means to understand and describe the significant single - spin asymmetries @xmath59 observed @xcite in @xmath60 , with the pion at high @xmath15 .
these are inclusive `` left - right '' asymmetries and may be generated by the sivers function from the effects of the quark intrinsic transverse momentum @xmath51 on the partonic hard - scattering which has a steep @xmath15 dependence .
the resulting asymmetry @xmath59 is then power - suppressed as @xmath61 in qcd , where @xmath62 is an average intrinsic transverse momentum .
similar effects may arise also from the collins function .
fits to the available @xmath59 data have been performed recently @xcite , assuming variously dominance of the collins or the sivers mechanisms .
an exciting new development in the field is that the star collaboration has presented the first data on @xmath63 from rhic @xcite .
the results are shown in fig .
as one can see , a large @xmath59 persists to these much higher energies .
[ fig9 ] also shows predictions based on the collins and the sivers effects @xcite , and on a formalism @xcite that systematically treats the power - suppression of @xmath59 in terms of higher - twist parton correlation functions ( for a connection of the latter with the sivers effect , see @xcite ) .
the star data clearly give valuable information already now .
for the future , it will be important to extend the measurements to higher @xmath15 where the perturbative - qcd framework underlying all calculations will become more reliable .
it was recognized some time ago that certain fourier transforms of generalized parton densities with respect to momentum transfer give information on the position space distributions of partons in the nucleon @xcite . for a transversely polarized nucleon
, one then expects @xcite a distortion of the parton distributions in the transverse plane , which could provide an intuitive physical picture for the origins of single - spin asymmetries .
we finally note that _
double_-transverse spin asymmetries @xmath64 in @xmath0 scattering offer another possibility to access transversity .
candidate processes are drell - yan , prompt photon , and jet production .
recently , the nlo corrections to @xmath65 have been calculated @xcite .
the results show that @xmath64 is expected rather small at rhic .
i am grateful to the organizers of dis 2003 for their invitation .
i thank d. boer , g. bunce , m. grosse - perdekamp , s. kretzer , z. meziani , g. rakness , m. stratmann for very useful discussions and help , and riken , brookhaven national laboratory and the u.s .
department of energy ( contract number de - ac02 - 98ch10886 ) for providing the facilities essential for the completion of this work .
0 for recent reviews , see : b. lampe , e. reya , phys .
* 332 * ( 2000 ) 1 ; + e.w .
hughes , r. voss , annu .
nucl . part .
* 49 * ( 1999 ) 303 ; + b.w .
filippone , x .- d .
ji , adv . nucl
* 26 * ( 2001 ) 1 .
u. stsslein , acta phys .
* b33 * ( 2002 ) 2813 .
bjorken , phys . rev .
* 148 * ( 1966 ) 1467 ; _ ibid . _
* d1 * ( 1970 ) 1376 . x .-
ji , phys .
* 78 * ( 1997 ) 610 . for information on the erhic project ,
see : http://www.bnl.gov/eic l.c .
bland , hep - ex/0212013 . c. marchand , these proceedings .
a. bravar , d. von harrach , a.m. kotzinian , phys .
* b421 * ( 1998 ) 349 ; a. airapetian et al .
, hermes collab .
lett . * 84 * ( 2000 ) 4047 .
see , for example : g. bunce , n. saito , j. soffer , and w. vogelsang , annu .
nucl . part .
sci . * 50 * , 525 ( 2000 ) , and references therein .
see : b. jger , a. schfer , m. stratmann , w. vogelsang , phys .
* d67 * ( 2003 ) 054005 , and references therein .
m. glck et al . , e. reya , m. stratmann , w. vogelsang , phys . rev .
* d63 * ( 2001 ) 094005 .
a. bazilevsky , talk presented at the `` x@xmath66 workshop on high energy spin physics ( spin-03 ) '' , dubna , russia , sep .
16 - 20 , 2003 .
adler et al . ,
phenix collab .
, hep - ex/0304038 . b. adeva et al . , smc , phys . lett .
* b420 * ( 1998 ) 180 .
a. airapetian et al .
, hermes collab .
, hep - ex/0307064 . j. blmlein , h. bttcher , nucl .
* b636 * ( 2002 ) 225 .
m. stratmann , w. vogelsang , phys .
* d64 * ( 2001 ) 114007 ; m. glck , e. reya , hep - ph/0203063 ; a. kotzinian , phys .
* b552 * ( 2003 ) 172 ; g. navarro , r. sassot , eur .
j. * c28 * ( 2003 ) 321 ; e. christova , s. kretzer , e. leader , eur .
j. * c22 * ( 2001 ) 269 ; e. leader , d.b .
stamenov , phys .
* d67 * ( 2003 ) 037503 ; s.d .
bass , phys .
* d67 * ( 2003 ) 097502 .
x. zheng et al .
, jefferson lab hall a collab .
, nucl - ex/0308011 . n. isgur , phys . rev .
* d59 * ( 1999 ) 034013 .
g. farrar , d.r .
jackson , phys .
* 35 * ( 1975 ) 1416 ; s.j .
brodsky , m. burkardt , i. schmidt , nucl .
* b441 * ( 1995 ) 197 ; see also : e. leader , a.v .
sidorov , d.b .
stamenov , int . j. mod
a13 * ( 1998 ) 5573 .
ralston , d.e .
soper , nucl .
* b152 * ( 1979 ) 109 ; x. artru , m. mekhfi , z. phys . * c45 * ( 1990 ) 669 ; r.l .
jaffe , x. ji , phys .
* 67 * ( 1991 ) 552 ; nucl . phys .
* b375 * ( 1992 ) 527 .
collins , nucl .
b394 * ( 1993 ) 169 .
collins , nucl .
* b396 * ( 1993 ) 161 .
a. airapetian et al . , hermes collab . ,
* 84 * ( 2000 ) 4047 .
oganessyan , h.r .
avakian , n. bianchi , a.m. kotzinian , hep - ph/9808368 ; p.j .
mulders , r.d tangerman , nucl .
b461 * ( 1996 ) 197 .
sivers , phys . rev . *
d41 * ( 1990 ) 83 ; _ ibid . _
* d43 * ( 1991 ) 261 .
brodsky , d.s .
hwang , i. schmidt , phys .
* b530 * ( 2002 ) 99 .
collins , phys .
* b536 * ( 2002 ) 43 .
belitsky , x. ji , f. yuan , nucl .
( 2003 ) 165 .
d. boer , p.j .
mulders , f. pijlman , nucl .
( 2003 ) 201 .
d. boer , p.j .
mulders , phys .
* d57 * ( 1998 ) 5780 ; + d. boer , phys .
* d60 * ( 1999 ) 014012 .
d. boer , r. jakob , p.j .
mulders , nucl .
b504 * ( 1997 ) 345 ; phys . lett .
* b424 * ( 1998 ) 143 ; d. boer , nucl .
* b603 * ( 2001 ) 195 ; k. kasuko et al . , belle collab . , talk at `` spin 2002 '' , aip conf .
675 , p. 454 .
a. metz , phys .
* b549 * ( 2002 ) 139 .
a. bacchetta , r. kundu , a. metz , p.j .
mulders , phys .
* b506 * ( 2001 ) 155 ; phys .
* d65 * ( 2002 ) 094021 ; a. bacchetta , a. metz , j.j .
yang , hep - ph/0307282 ; l.p .
gamberg , g.r .
goldstein , k.a .
oganessyan , phys .
* d67 * ( 2003 ) 071504 ; phys .
* d68 * ( 2003 ) 051501 ; hep - ph/0309137 .
see , for example , d.l .
adams et al . , e704 collab . , phys . lett .
* b261 * ( 1991 ) 201 ; * b264 * ( 1991 ) 462 ; a. bravar et al .
* 77 * ( 1996 ) 2626 .
m. anselmino , m. boglione , f. murgia , phys .
* d60 * ( 1999 ) 054027 ; m. boglione , e. leader , phys .
* d61 * ( 2000 ) 114001 ; u. dalesio , f. murgia , hep - ph/0211454 .
qiu , g. sterman , phys . rev .
* d59 * ( 1998 ) 014004 . y. koike , nucl .
* a721 * ( 2003 ) 364 ; hep - ph/0210396 .
m. burkardt , int . j. mod
a18 * ( 2003 ) 173 , and references therein ; m. diehl , eur .
j. * c25 * ( 2002 ) 223 ; for a recent different approach , see : a.v .
belitsky , x. ji , f. yuan , hep - ph/0307383 .
m. burkardt , phys .
* d66 * ( 2002 ) 114005 ; hep - ph/0302144 ; m. burkardt , d.s .
hwang , hep - ph/0309072 .
a. mukherjee , m. stratmann , w. vogelsang , phys .
* d67 * ( 2003 ) 114006 . | we briefly review some of the recent developments in qcd spin physics .
bnl - nt-03/28 + rbrc-338 + |
heavy ion collisions have received significant attention in recent years . electromagnetic probes ( photons , dileptons etc )
have been proposed to be one of the most promising tools to characterize the initial state of the collisions @xcite . because of the very nature of their interactions with the constituents of the system they tend to leave the system almost unscattered .
photons are produced at various stages of the evolution process .
the initial hard scatterings ( compton and annihilation ) of partons lead to photon production which we call hard photons .
if quark gluon plasma ( qgp ) is produced initially , there are qgp - photons from thermal compton plus annihilation processes .
photons are also produced from different hadronic reactions from hadronic matter either formed initially ( no qgp scenario ) or realized as a result of a phase transition from qgp .
these apart , there exits another class of photon emission process via the jet conversion mechanism ( jet - plasma interaction ) @xcite which occurs when a high energy jet interacts with the medium constituents via annihilation and compton processes . in current heavy ion collision experiments ,
the temperature @xmath1 is not only the important scale , momentum scale , @xmath2 , ( of the partons ) is also important .
therefore running of the coupling in the high momentum regime ( @xmath3 ) has to be taken into account to calculate the cross sections and the energy - loss processes . in this work
we calculate photons from jet - plasma interaction taking into account running of qcd coupling and both collisional and radiative energy losses .
the plan of the article is as follows .
we discuss the formalism in the next section .
results will be discussed in the section 3 .
finally we will conclude .
the lowest order processes for photon emission from qgp are the compton scattering ( @xmath4 ) and annihilation ( @xmath5 ) process .
the differential photon production rate for this process is given by @xcite : @xmath6 where , @xmath7 represents the spin averaged matrix element squared for one of those processes which contributes in the photon rate and @xmath8 is the degeneracy factor of the corresponding process .
@xmath9 , @xmath10 and @xmath11 are the initial state and final state partons . @xmath10 and @xmath11 are the bose - einstein or fermi - dirac distribution functions .
in the photon production rate ( from jet - plasma interaction ) one of the collision partners is assumed to be in equilibrium and the other ( the jet ) is executing random motion in the heat bath provided by quarks ( anti - quarks ) and gluons . furthermore , the interaction of the jet is dominated by small angle scattering . in such scenario the evolution of the jet phase space distribution
is governed by fokker - planck ( fp ) equation where the collision integral is approximated by appropriately defined drag ( @xmath12 ) and diffusion coefficients @xcite .
the drag and diffusion coefficients are infrared singular .
the infra - red cut - off is fixed by plasma effects , where only the medium part is considered , completely neglecting the vacuum contribution leading to ambiguity in the energy loss calculation . if the latter part is taken into account the strong coupling should be running .
thus for any consistent calculation one has to take into consideration this fact . in that case @xmath13 ( @xmath14 in this case ) , and the above integrals must be evaluated numerically where the infra - red cut - off is fixed by debye mass to be solved self - consistently : @xmath15 here the strong coupling which we take as running , i. e. @xmath16 .
we chose the following parametrization of @xmath0 which respects the perturbative ultra - violet ( uv ) behavior and the 3d infra - red ( ir ) point @xcite : @xmath17 with @xmath18 in this case .
the parameters @xmath19 , @xmath20 and @xmath21 are given by @xmath22 , @xmath23 and @xmath24 gev .
for the limiting behavior ( @xmath25 ) of the coupling we choose , @xmath26 here @xmath27 and @xmath28 denote the values of the ir fixed point of @xmath29 yang - mills theory in @xmath30 and @xmath31 dimensions , respectively .
the remaining four parameters ( @xmath32 and @xmath33 ) fit the numerical results for pure yang - mills theory obtained from the rg equations in ref .
@xcite . in our calculation
we have considered both collisional and radiative energy losses in the following manner .
@xmath34\end{aligned}\ ] ] for running @xmath0 , the expressions for the collisional and radiative energy losses can be found in @xcite .
having known the drag and diffusion , we solve the fp equation using green s function techniques ( for details see ref .
@xcite ) . in order to obtain the space - time integrated rate we first note that the phase space distribution function for the incoming jet in the mid rapidity region is given by
( see ref .
@xcite for details ) @xmath35 with this jet parton phase space distribution function one can easily obtain jet photon yield from eqn .
( 1 ) : @xmath36
in order to obtain the photon @xmath37 distribution we numerically integrate eq .
( [ last ] ) .
the results for jet - photons for rhic energies are plotted in fig .
[ fig_rhic446 ] ( left ) where we have taken @xmath38 mev and @xmath39 fm / c .
we find that the yield is decreased with the inclusion of both the energy loss mechanisms as compared to the case when only collisional energy loss is considered .
it is to be noted that when one considers collisional energy loss alone the yield with constant @xmath0 is more compared to the situation when running @xmath0 is taken into account ( see fig .
[ fig_rhic446 ] left ) . in order to compare our results with high @xmath37 photon data measured by the phenix collaboration @xcite
, we have to evaluate the contributions to the photons from other sources , that might contribute in this @xmath37 range . in fig . [ fig_rhic446 ] ( right )
the results for jet - photons corresponding to the rhic energies are shown , where we have taken @xmath38 mev and @xmath39 fm / c .
the individual contributions from hard and bremsstrahlung processes @xcite are also shown for comparison .
we have calculated the transverse momentum distribution of photons from jet plasma interaction with running coupling , i. e. with @xmath40 where we have included both collisional and radiative energy losses .
using running coupling we find that the depletion in the photon @xmath37 spectra is by a factor of @xmath41 more as compared to the case with constant coupling for rhic energies phenix photon data have been contrasted with the present calculation and the data seem to have been reproduced well in the low @xmath37 domain .
the energy of the jet quark to produce photons in this range ( @xmath42 ) is such that collisional energy loss plays important role here .
it is shown that inclusion of radiative energy loss also describes the data reasonable well . | we calculate photons from jet - plasma interaction considering collisional and radiative energy loss of jet parton .
the phase space distribution of the participating jet is dynamically evolved by solving fokker - planck equation .
we treat the strong coupling constant ( @xmath0 ) as function of momentum and temperature while calculating the drag and diffusion coefficients .
it is observed that the quenching factor is substantially modified as compared to the case when @xmath0 is taken as constant .
it is shown that the phenix data is reasonably well reproduced when contributions from all the relevant sources are taken into account .
energy loss , running coupling , qgp |
the ` knee ' , a rather sharp steepening in the primary cosmic ray ( cr ) energy spectrum at about 3 pev , was inferred from the observation of a similar feature in the measured size spectrum of extensive air showers by kulikov and khristiansen ( 1958 ) .
the knee is commonly asserted to be due to an increasing failure of ` galactic containment ' of the cr generated by sources within the galaxy , the containment being caused by the magnetic fields in the interstellar medium ( ism ) , however , it is the firmly - held view of the present authors that the knee is too sharp for this explanation and we have advanced what we claim to be a more realistic model .
this is our ` single source ( ss ) model ' ( see erlykin and wolfendale , 1997 , 2001b for recent details ) which comprises cosmic ray acceleration up to the knee energy by supernova remnants , the knee itself being due to the truncation that occurs at 3 pev for oxygen nuclei from a single , recent , nearby snr .
the other main accelerated nucleus at these energies is iron and its termination occurs at about 12 pev where , it is claimed by us that there is a small second knee ( when the spectrum is plotted as @xmath6 vs. @xmath7 , the knees appear as small peaks ) . the remainder of the cr spectrum ( at least to some 10@xmath8 gev , or so ) is presumed due to ` super'-snr and other sources and their spatial distribution is such as to give a comparatively smooth spectrum in the pev region .
erlykin and wolfendale ( to be referred to henceforth as ew ) have examined a variety of other cosmic ray data and concluded that there is either support for the model or that the data are neutral .
very recently , low energy gamma ray data have also been studied ( ew , 2002 ) and the well - known ` gamma ray excess ' in the inner galaxy , and deficit in the outer galaxy , have been explained in terms of propagation differences dependent on the conditions in the ism from which the gamma rays come .
the results relate to galaxy - wide properties and , although the snr acceleration hypothesis has been invoked , there is no significant information about the single source .
it is at higher gamma ray energies where potential problems exist ( e.g. drury et al . , 1994 ) .
most recently , bhadra ( 2002 ) has argued that the single source should be visible in tev gamma rays , and it is not .
this is the topic to be addressed here .
we use the results found in a very recent paper ( ew , 2003a to be referred to as i ) , where we made predictions of the fluxes and the angular distribution of gamma ray intensity from snr of different ages and at different distances from the sun .
the threshold energies were taken as 0.1 gev and 1 tev .
a critical feature of the bhadra estimate was the ` normalisation ' of the snr ` conditions ' so as to give the cr energy density created by the single source at earth .
we regard this as a legitimate procedure and we follow this path , although other features of our model differ considerably from those adopted by bhadra .
our calculations are thus not simply a ` re - run of the bhadra calculations with different values for the parameters ' but , rather , for what is certainly a more appropriate model of snr acceleration and ( less certainly , perhaps ) a significantly different model of cosmic ray propagation .
bhadra s model is rather straightforward , in principle , at least : particles are accelerated by the snr shock to give a differential spectrum @xmath9 , with @xmath10 and these particles interact with the ambient ( or swept - up ) gas of density @xmath11 , with @xmath12 @xmath13 . at this stage
it is necessary to make critical remarks , however .
implicit in the bhadra calculations is the assumption that the snr shock accelerates the cr instantaneously , at ` @xmath14 ' .
although this can be used to give a viable mathematical model , such a situation is certainly not appropriate to a real snr where the acceleration occurs over an extended period : @xmath15 years in our model , and little different in other snr acceleration models .
nevertheless , we continue to describe the bhadra calculations .
it is assumed that the particles diffuse from the source a distance @xmath16 from the sun in a normal , gaussian fashion with diffusion coefficient @xmath17 @xmath18s@xmath19 .
the parameters are chosen to give the required cr energy density at the earth created by the single source .
for instance , if the source is at the distance of 300 pc and it is 10@xmath20 years old the total energy transferred from sn to cr is required to be 1.9@xmath2110@xmath22 erg .
the table summarizes the most important parameters . .
values of the parameters adopted by bhadra ( 2002 ) in comparison with those in the present work ( denoted ew ) .
the remnant is taken to be at 300 pc from the sun .
@xmath23 is the cosmic ray energy input from the snr ; @xmath24 is the diffusion coefficient for normal , gaussian diffusion , @xmath7 is in gev ; @xmath11 is the density of the interstellar medium ( ism ) in _
h_-atoms @xmath13 ; @xmath25 is the exponent of the differential proton spectrum ; @xmath26 and @xmath27 are the predicted gamma ray fluxes . [ cols="<,^,^ " , ] the expected minimum gamma ray flux above 0.1 gev rises with the age of the sn from @xmath28 at 10@xmath29 years to @xmath30 at 10@xmath31 years and from @xmath32 to @xmath33 above 1 tev , respectively .
for the real candidates for the single source discussed in ew , ( 1997 ) ( eg . loop i , clayton snr ) the expected fluxes are substantially higher . comparing these fluxes with the diffuse gamma ray background bhadra found that for the present gamma ray telescopes it should have been possible to observe the single source .
since there has been no claim for an observation bhadra concluded that the single source can not be such a snr .
turning to our remarks on the validity of bhadra s model , in addition to the basic problem with the assumption about the instantaneous acceleration there are two further reasons why we can not allow this conclusion to stand and , in fact , bhadra made the neeeded reservations , viz .
_ the detection could be crucial , depending on the angular size of the object _ and _ unless the source is in a lower density environment_. these are the points addressed in the following sections .
we require , first , the likely whereabouts of the single source such that it can give the particle spectrum needed in the knee region .
we have calculated cosmic ray energy spectra originating from snr of different ages and distances .
the model of acceleration was described in ew ( 2001a ) and briefly in i. the propagation of accelerated cosmic rays through the ism was calculated using two alternative assumptions about the mode of propagation from the source : ` anomalous ' diffusion , viz . making allowance for the fractal - like nature of the ism , and normal , gaussian , diffusion . following the work of lagutin et al.(2001a ,
b ) we distinguish these two modes by a parameter @xmath34 determined by the fractal nature of the medium ; @xmath34 is equal to 1 or 2 for anomalous or normal diffusion , respectively .
the difference between the two modes can be seen most clearly in the shape of the lateral distribution function for the cosmic ray intensity : @xmath35 for @xmath36 and @xmath37 for @xmath38 , with @xmath39 , @xmath16 being the distance from the radius @xmath40 pc where the particles start to diffuse and @xmath41 being the diffusion radius which is defined as @xmath42 , i.e. , there is a different time dependence for the two modes .
kpc for the vertical scale of the galactic halo and @xmath44 and @xmath45 are the diffusion time and time against escape , respectively , for the protons .
details of the propagation model are given in ew ( 2002 ) and in i. the difference in the two lateral distributions is quite dramatic .
thus , for @xmath46 2 , 4 , 6 and 8 , the ratio of the intensity for anomalous diffusion to that for normal diffusion changes as 0.11 , 0.19 , 5.9 and 2.1@xmath2110@xmath29 , respectively . the long tail for anomalous diffusion - the occasional considerable ` penetration ' ( levy flights ) - can have important implications .
an example is that for the secondary to primary ratio ; to our knowledge the implications have not been worked out .
we have determined the proton energy spectra expected for the two values of @xmath34 and these are given in figures 1 and 2 .
also indicated in the figures is the spectrum ` required ' by the ss model .
we have argued that the ` needed ' particles are oxygen nuclei for the first ` peak ' at 3 pev and iron nuclei for the second ` peak ' at @xmath47 pev , and the requirement has been converted to rigidity before plotting , in figures 1 and 2 .
an alternative association of the ` peaks ' with helium and oxygen changes the indicated ss spectrum no more than by 17% .
( ew , 2003b ) .
dotted lines above and below the ` ss ' curve indicate its uncertainty limits ; the least uncertainty and therefore the most important constraint of the ss model is in the knee region .
the limits come from the fit to the experimental data on the primary energy spectrum measured by means of the cherenkov light emitted by extensive air showers ( ew , 2001b ) .
these data determine the magnitude of the uncertainty in the range of about a decade below the knee . at lower energies ,
- @xmath48 gev - the upper limit is determined by the uncertainty of the direct measurements of the primary cr energy spectrum ( biermann and wiebel - sooth , 1999 ) .
although the shape of the ss spectrum has been adopted from the theoretical model of berezhko et al .
( 1996 ) its lower experimental limit at low energies is completely uncertain because the contribution of the single source to the total cr intensity at these energies is negligibly small and even consistent with zero .
however , the actual value of the uncertainty at tev energies is not important for this analysis because we have made a quantitative comparison of the calculated proton spectra with the ss spectrum only in the most important range of energies , viz .
that covering a decade below the knee .
the energy range of @xmath49 used for this comparison is indicated in the figures by _ ` min ' _ and _ ` max'_. ( ew , 2002 ) ( rigidity and energy are , of course , the same for protons ) .
the energy interval used for the comparison of proton spectra with the ss model is marked by _ min _ and _ max_. the result of the comparison in terms of the reduced @xmath50 , i.e. @xmath50 divided by the number of degrees of freedom , @xmath51 , is shown inside the figures.,width=604,height=529 ] [ fig : bhad1 ] .,width=604,height=529 ] [ fig : bhad2 ]
some comments are needed about the figures .
it is evident that @xmath52 always gives a bad fit ( figure 2 ) whereas @xmath53 can give a reasonable one , at least for energies above 10@xmath20 gev , up to the cut - off at 4@xmath54 gev , for a range of age ( t ) and distance ( r ) values .
has a formally acceptable value for @xmath38 at t@xmath55 kyear the slopes of the proton spectra are too steep to give the sharp knee and we do not consider it as a good contender . ]
calculations made for a wide range of t and r allow us to estimate the range over which there is satisfactory agreement between calculations and the ss model ; it is @xmath56 kyear and @xmath57 pc for the adopted set of input parameters .
it is relevant to point out that larger distances , too , would give a good fit to the spectral _
shape_. the necessary upward movement in intensity could be effected by increasing the fraction of the shock energy going into cr . as was pointed out in i , berezhko et al.,(1996 ) used much bigger values than the 10% used here - their highest being 80% .
specifically , for 100 kyear an increase in cr yield by a factor 8 would allow the source to be at @xmath58 pc . the shape and the absolute intensity of the cr energy spectrum
give the most stringent constraints on the age and distance .
the t - r region of snr which could give an acceptable spectrum ( for our standard ` 10% ' ) is shown in figure 3 by the 95% confidence level contour .
following bhadra s approach we have also used such an integrated characteristic of the spectrum as the energy density contained in it .
again , we used for the comparison just the last decade of the spectrum below the knee , because intensities at low energies are so poorly determined by the single source model .
the energy density contained in the spectrum of our single source between @xmath59 gev and @xmath60 gev is 1.84@xmath61 .
comparison of this value , allowing for uncertainty , with those expected for snr of different ages and distances gives an acceptable t - r region indicated by the dashed line in figure 3 .
it overlaps with the region deduced from the comparison of the spectral shapes .
this proves the consistency of these results , although we must admit that the two methods are not completely independent , because consistent values for the spectra should inevitably give consistent energy densities contained in them .
in any case this analysis indicates that our single source should be located at about 300 - 350 pc from the sun and should be about 90 - 100 kyears old .
we know of no objection to such parameters ( see 4 for its likely location ) .
s@xmath19 . ,
width=566,height=264 ] [ fig : bhad3 ] fluxes of gamma rays from snr of different ages and at different distances from the sun have been calculated in i. in figure 3 we show contours of the fluxes calculated for the interactions of protons , accelerated by the snr , which propagate through the ism by way of anomalous diffusion ( @xmath36 ) and collide with nuclei of the ism , the density of the ism for these calculations has been taken to be our standard @xmath62 .
it is seen that the expected fluxes are about 3@xmath63 for @xmath64 gev and about 2@xmath65 for @xmath66 tev .
it is necessary now to study the implications of the fact that the cosmic rays from our single sourse are not only protons , but that the composition is mixed . at the same rigidity
it consists of 21% p , 48% o , 13% ` heavy ' nuclei ( @xmath67 ) and 18% fe .
because nuclei of the same rigidity are more efficient in the production of gamma quanta ( see i ) , the expected fluxes should be increased by a factor of 10.5 .
this gives expected fluxes of @xmath68 for @xmath64 gev and @xmath69 for @xmath66 tev .
it is necessary to point out that the cr energy injection is higher by the same amount . since this is a first order effect , a detailed examination of the problem
is required .
as pointed out in i , there are several aspects that need attention , specifically : + ( i ) the effect of the pre - snr winds from the progenitor star in excavating a ` hole ' into which the snr shock expands , this is essentially the origin of the hism ( note that for vey young snr , however , the progenitor star s wind may have _ enhanced _ the density ) ; + ( ii ) the general ism density in the region where the progenitor star was situated and + ( iii ) the pressure , or otherwise , of clumpy gas ( often molecular ) in the vicinity of the snr . in the present case ,
it is almost certain that the progenitor star - at 300 - 350 pc from the sun - was in the hot interstellar medium , where the density is often quoted as @xmath70 @xmath13 ( berezhko et al . , 1996 ;
cox and reynolds , 1987 , for example quote a density of @xmath71@xmath13 ) . indeed ,
since all our model calculations assume this to be the case , a low density target material is a prerequisite .
the assumption of the hism comes from two factors : + ( i ) the absolute maximum particle energy needed to explain the knee ( @xmath72 gev for oxygen nuclei ) appears only for this density , in the berezhko et al .
model , and + ( ii ) the hism is eminently reasonable for a nearby source . factor ( ii )
can be examined in more detail .
insofar as the sun is located on the edge of a spiral arm , in one hemisphere ( the south ) , at least , the gas density will be low .
frish ( 1997 ) has considered the situation in detail . beyond the very local region , where there is the ` local fluff ' ( of extent @xmath733 pc ) in certain directions the column density of atomic ( and molecular ) hydrogen is very low .
the ` local bubble ' ( loop i ) is quickly reached , this bubble being caused by several sn over the past myear . here
, one expects the hism , with its density @xmath74 @xmath13 .
concerning the interarm region , frish quotes an undisturbed part as having a density of @xmath75 @xmath13 , i.e. even smaller .
however , much of the local bubble is here and the density will be higher because of material brought in from elsewhere .
interestingly , frish ( 1981 ) suggested , earlier , that the sun is embedded in one of the super - bubble shells associated with the formation of the scorpius - centaurus association .
another possibility for the formation of loop
i ( but very similar to that given above is that it was caused by activity in the ` upper centaurus lupus ' subgroup some 14 - 15 myear ago ( de geus , 1991 ) ) .
a distinction must be made between a ` recent ' sn exploding in loop i , i.e. in the low density hism caused by previous snr and stellar activity , and the low level gamma ray flux from loop i as a whole .
this latter was considered in i and it was argued there that the mean density overall , allowing for molecular clouds inside the loop , and the piled - up gas in the edges of the loop , is @xmath76 0.1 @xmath13 .
there is no conflict with our @xmath77 @xmath13 if the single source is well into the interior . with @xmath78 ,
the predicted flux is reduced to @xmath79 @xmath80s@xmath19 for @xmath810.1 gev and @xmath82@xmath80s@xmath19 for @xmath83 1 tev .
if , surprisingly , the source is not in the hism in the local bubble , but is isolated , then as was discussed in i , @xmath84 and the predicted fluxes are @xmath85 for @xmath64 gev and @xmath86 for @xmath66 tev . the predicted fluxes are shown in figure 4 , for the various possibilities of _ n _ and _
z_. gev and @xmath87 tev for a variety of scenarios .
the source would be at @xmath88 pc from us and have an age of @xmath89 kyear , i.e. it would be only 10 - 20 kyear since the particles were released by the remnant .
the dotted lines marked ` needed ' are the minimum fluxes required for the source to have been detected by the arrays in use up to now . as an indication of the future
we also give the estimated minimum detectable flux for @xmath66 tev for a source of 20@xmath1 radius , from the work of aharonian et al.,(1997 ) .
these authors give results for 1000 hours of observation of a point source and one of 1@xmath1 extent ; our estimate arises from an extrapolation , based on results described in i. the work reported relates to the then proposed iact ( 100 gev - class imaging atmospheric telescope array ) , denoted ` hess'.,width=566,height=340 ] [ fig : bhad4 ] the sources must be detected against a background due to the galactic diffuse emission arising from cr - ism nucleus interactions ( and , in the case of tev gamma rays , protons interacting in the atmosphere ) .
this background can be allowed for in a straightforward manner for ` point ' sources , where a subtraction can be made of the signal nearby to the source , but for extended sources the problem is much more severe .
the angular radius of a snr , which is 90 - 100 kyear old , seen from a distance of 300 - 350 pc is about 20@xmath1 ( see also figure 5 in i ) .
for such a large size the determination of the background by a linear interpolation of the intensity between the adjacent regions is not at all accurate .
estimates of the background based on the known column density of the target gas , although more appropriate , are again not sufficiently accurate because the intensity of the initiating cosmic rays along the line of sight can not be assumed to be strictly constant
. estimates of the limiting fluxes were made from the available data and they are given in i. for a source of radius 20@xmath1 the average limits are @xmath90 for @xmath64 gev and @xmath91 for @xmath66 tev .
these limiting sensitivities , which relate to observations made so far , are indicated in figure 4 .
the lower energy limit comes from an extrapolation of the egret results ( hartman et al , 1999 ) , which refer to small angular sizes , to 20@xmath1 .
that for the upper energy band comes largely from the tibet and hegra arrays ( amenomori et al . ,
2001 and lampeitl et al . , 2001 ,
respectively ) .
it is instructive to examine these sensitivities in terms of the background fluxes .
for @xmath64 gev , the intensity at low latitudes towards the galactic anticenter is about @xmath92 and for our 20@xmath1-radius source ( @xmath93 = 0.4 sr ) the flux would be @xmath94 .
the plotted limit of @xmath95 is thus 8% of the background . for the direction to the galactic center , where the background is higher by a factor of 3 ,
this percentage is reduced to ( 2 - 3)% .
when allowance is made for other sources of uncertainty our adopted limit seems reasonable . for @xmath66 tev
, the corresponding background in the galactic anticenter direction has been estimated to be @xmath96 ( porter and protheroe , 1999 ) , viz a flux of @xmath97 over a 20@xmath1-radius source .
our adopted limit of @xmath98 is thus the same : @xmath738% of the background . in the direction of the galactic center
the background is expected to be twice as large and the fraction of the limiting flux falls to @xmath734% .
this figure , too , is reasonable in view of misidentified proton contributions and a variety of technical problems associated with making absolute measurements at different zenith angles .
the conclusions are indicated in figure 4
. only if the single source is in a ` high ' density region ( @xmath99 ) and the primaries are ` heavy ' nuclei ( z = 10 ) ( with consequent high cr energy injection ) will it be possible to detect the single source at high energies .
such a situation is not impossible but there would be problems for our single source model with such a high ambient density , specifically that it would not be possible with the model in use to reach the required 3 pev energy at the knee .
insofar as we consider it very likely that the source is in the hism and is extended , the chance of detecting it with contemporary instruments is considered , by us , to be very low .
there are , however , hopes for the future .
it is germane to consider in which direction future , improved , gamma ray detectors should be pointed in order to see it .
the best that can be done here is to suggest the general direction of loop i. however , since this structure occupies about 25% of the sky the ` advice ' is not very helpful .
the authors are grateful to the royal society and the university of durham for financial support .
paula chadwick , of the hess collaboration , is thanked for providing useful information .
two unknown referees are also thanked for useful comments and suggestions .
aharonian , f.a .
et al . , 1997 , _ astroparticle physics _ , * 6 * , 369 + amenomori , m. et al . , 2001 , _
_ ( hamburg ) , * 6 * , 2344 + berezhko , e.g. , elshin , v.k . and ksenofontov , l.t . , 1996 ,
_ , * 82 * , 1 + bhadra , a. , 2002 , _ j. phys .
g : nucl . part .
_ , * 28 * , 397 + biermann p.l . and wiebel - sooth b. , 1999 , _ astronomy & astrophysics - interstellar matter , galaxy , universe _ , landolt - brnstein , springer verlag , berlin / heidelberg , * 3 * , 37 + cox d.p . and reynolds r.j . , 1987 , _ annual rev .
astrophys . _ * 25 * , 303 + de geus , e.j . , 1991 , _
_ , * 262 * , 258 + drury , l.oc . , aharonian , f.a . and vlk , 1994 , _ astron .
_ , * 287 * , 959 + erlykin , a.d . and wolfendale , a.w . , 1997 ,
_ j. phys .
g : nucl . part .
_ , * 23 * , 979 + erlykin , a.d . and
wolfendale , a.w . , 2001a , _ j. phys .
g : nucl . part .
_ , * 27 * , 941 + erlykin , a.d . and
wolfendale , a.w . , 2001b , _ j. phys .
_ , * 27 * , 1005 + erlykin , a.d . and wolfendale , a.w . , 2002 ,
_ j. phys .
g : nucl . part .
_ , * 28 * , 2329 + erlykin , a.d . and wolfendale , a.w . ,
2003a ( submitted to j. phys .
g ) + erlykin , a.d . and
wolfendale , a.w .
, 2003b , _ nucl .
b ( proc .
suppl . ) _
( in press ) + frisch , p.c . , 1981 , _ nature _ , * 293 * , 377 + frisch , p.c .
, 1997 , astro - ph/9705231 + hartman , r.c .
et al . , 1999 , _
_ , * 123 * , 179 + kulikov , g.v . and khristiansen , g.b .
, 1958 , _ j. exp
_ , * 35 * , 635 + lagutin , a.a .
2001a , _ nucl .
b ( proc .
suppl . )
_ , * 97 * , 267 + lagutin , a.a .
et al . 2001b , _ proc .
_ ( hamburg ) , * 5 * , 1900 + lampeitl , h. et al . , 2001 , _ 27th int . cosm
. ray conf . _
( hamburg ) , * 6 * , 2348 + porter t.a . and protheroe r.j . , 1999 , _ 26th int .
_ , salt lake city , * 4*. 306 + | some six years ago , we ( erlykin and wolfendale , 1997 ) proposed the ` single source model ' in which a local , recent supernova remnant ( snr ) was responsible for the ` knee ' in the cosmic ray ( cr ) energy spectrum at @xmath0 pev . stimulated by the paper by bhadra ( 2002 ) , which drew attention to a possible gamma ray signature of this local remnant , we now study the situation for the local source and we conclude that , in contrast to bhadra s conclusion , the non - observation of this remnant is understandable - at least using our snr model .
it is due to the fact that this snr , being local , develops in the local hot interstellar medium ( hism ) with its low density of gas and also being nearby it will be an extended source occupying up to 40@xmath1 of the sky and thus indistinguishable from the background . plus 2 mm minus 2 mm 23.0 cm 17.0 cm -1.0 in -42pt * high - energy cosmic gamma rays + from the ` single source ' * ' '' '' a.d . erlykin@xmath2 and a.w .
wolfendale@xmath3 @xmath4 lebedev physical institute , leninsky prospekt , moscow , russia + @xmath5 department of physics , university of durham , dh1 3le , uk + |
we calculate the probability that an observed shower , containing @xmath70 muon decays , was initiated by a neutrino @xmath8 , of definite flavor @xmath71 , or @xmath72 .
the main observable of a shower is its energy @xmath68 , which is proportional to the total collected light . because the detector energy resolution is narrow ,
we simply take it to be flat in the range @xmath73 e_\text{sh}$ ] .
using bayes theorem , the probability that a shower with energy @xmath68 and @xmath70 muon decays was initiated by a @xmath8 is @xmath74 } \;.\ ] ] here , @xmath75 is the probability that a shower with energy @xmath68 is produced by the cc interaction of a @xmath8 , which we detail below , while @xmath76 is the probability that said shower yields @xmath70 muon decays , which is calculated via ` fluka ` simulations and shown in figs .
[ fig : muon_decay_distribution ] and [ fig : muon_decay_energies ] for different shower energies .
the probability @xmath77 is defined as @xmath78 where @xmath79 is the number of @xmath8-initiated showers generated by cc interactions .
the denominator in eq .
( [ equ : probnuadef ] ) is the total number of showers initiated by all flavors of neutrinos and anti - neutrinos . to calculate the number of showers
, we use the `` theorist s approach '' @xcite , assuming perfect detector efficiency at the relevant energies .
the final results on flavor discrimination are affected by only the relative , not the absolute , event rates from different flavors .
we consider a flux @xmath80 of @xmath8 ( in units of gev@xmath81 @xmath82 s@xmath81 sr@xmath81 ) arriving at the detector , which contains @xmath83 target nucleons .
the flux already includes any attenuation due to propagation in the earth . in observation time @xmath84 with detection solid angle @xmath85 ,
the number of detected @xmath8-initiated cc showers is @xmath86 where @xmath87 is the neutrino energy and @xmath88 is the neutrino - nucleon cc cross section @xcite .
the function @xmath89 is the probability that a neutrino with energy @xmath87 creates a shower with energy @xmath68 ; it is different for each flavor . * in @xmath0 cc interactions ,
all of the neutrino energy is deposited in the electromagnetic and hadronic showers .
accordingly , we define @xmath90 e_\text{sh } \\ & 0 , & & \text{otherwise } \end{aligned } \right . \;.\ ] ] * in @xmath1 cc interactions , the outgoing tau has numerous decay modes .
all of them have outgoing neutrinos , which carry away energy and do not appear in the shower , so that @xmath91 . on average ,
the outgoing neutrinos carry away 40% of the tau energy , or 25% of the primary neutrino energy . for simplicity , we make @xmath92 nonzero only in the energy range @xmath93 e_\text{sh } / 0.75 $ ] .
since 17% of tau decays are into muons and neutrinos , without a shower , we estimate @xmath94 e_\text{sh}/0.75 \\ & 0 , & & \text{otherwise } \end{aligned } \right . \;.\ ] ] * in nc interactions , the energy deposited in the shower is the energy of the final - state hadrons , _
i.e. _ , @xmath95 . for the shower energy to lie within 10% of @xmath68 , the value of @xmath18 must lie in the range @xmath96 \equiv [ 0.9,1.1 ] e_\text{sh}/e_\nu$ ] .
hence , we define @xmath97 where @xmath98 is the @xmath18 probability distribution for nc interactions @xcite . however , because hadron - initiated showers carry a small fraction @xmath18 of the neutrino energy , and because the neutrino flux is steeply falling , nc showers are subdominant to cc showers @xcite . * in @xmath23 cc interactions , the outgoing muon leaves an identifiable track .
we exclude these events by setting @xmath99 we have assumed that no track is mis - identified as a shower ; otherwise , the value of @xmath100 would be set to the probability of mis - identification . as with nc events , these would be subdominant in the shower spectrum .
we write eqs .
( [ equ : probflav1])([equ : numberofshowers1 ] ) in a more useful way .
consider an all - flavor astrophysical neutrino flux @xmath101 and flavor ratios at earth @xmath102 , such that the flux of @xmath8 is @xmath103 , with @xmath104 the normalization of the flux . with this , eq .
( [ equ : numberofshowers1 ] ) becomes @xmath105 with the shorthand @xmath106 finally , using eqs .
( [ equ : numberofshowers2 ] ) and ( [ equ : integraldef ] ) , and assuming equal flavor ratios for neutrinos and anti - neutrinos , eq .
( [ equ : probflav1 ] ) becomes @xmath107 } { \sum_{\alpha = e,\mu,\tau } f_{\alpha,\oplus } \left [ \left ( p_{n_\mu|\nu_\alpha}^\text{cc } i_{\nu_\alpha}^\text{cc } + p_{n_\mu|\bar{\nu}_\alpha}^\text{cc } i_{\bar{\nu}_\alpha}^\text{cc } \right ) + \left ( p_{n_\mu|\nu_\alpha}^\text{nc } i_{\nu_\alpha}^\text{nc } + p_{n_\mu|\bar{\nu}_\alpha}^\text{nc } i_{\bar{\nu}_\alpha}^\text{nc } \right ) \right ] } \;.\ ] ] the probability that the shower with @xmath70 muon decays was created by a @xmath8 or a @xmath47 is simply @xmath108 . , reflecting the central value and width of the band in fig .
[ fig : flavor_triangle ] . ] , reflecting the central value and width of the band in fig .
[ fig : flavor_triangle ] . ]
figure [ fig : nu_flavor ] shows this probability computed at @xmath41 tev , assuming a diffuse astrophysical neutrino flux with spectral index @xmath109 and a flavor composition of @xmath6 , compatible with icecube results @xcite .
the neutrino is more likely to be a @xmath0 if there are fewer muon decays and a @xmath1 if there are more decays .
the probability that the shower is from a @xmath23 nc interaction ( not shown ) reaches at most 10% , at large values of @xmath70 .
we use the results from appendix [ sec : formalism ] to infer the @xmath48 and @xmath49 flavor ratios of an ensemble of showers .
we first explain how we generate the artificial shower ensemble ; then we show how to infer their flavor ratios . to generate an ensemble of showers with energy @xmath68 , we first assume a neutrino flux with spectral index @xmath109 and `` real '' values for the flavor ratios @xmath110 .
we then use the probability distribution functions of the number of muon decays for each channel , @xmath111 and @xmath112 ( shown in figs .
[ fig : muon_decay_distribution ] and [ fig : muon_decay_energies ] ) , to construct the total probability distribution of muon decays associated to that flux , by summing over all flavors and interaction channels : @xmath113 \nonumber + \left [ \nu_\alpha \to \bar{\nu}_\alpha \right ] \;.
\end{aligned}\ ] ] figure [ fig : parameter_fit ] shows the total muon decay distribution for @xmath41 tev , for three choices of flavor composition .
the distribution for our nominal case @xmath6 has a saddle shape , peaked at low number of decays due to the sharp distribution of the @xmath0 cc channel .
the height of this peak increases with @xmath48 .
we use the above distribution to randomly sample the number of muon decays for each shower , that is , we obtain @xmath114 for @xmath115 .
these are our `` real '' data .
we choose @xmath116 , consistent with near - future expectations for icecube . merely to illustrate the flavor separation power given this sample size , we include binned data for this choice in fig .
[ fig : parameter_fit ] .
the points and error bars show the expected number of showers per bin and its @xmath117 poissonian fluctuation .
the power of flavor discrimination using muon echoes hinges on the ratio of the number of showers with few muon decays _
e.g. _ , @xmath118 to the number of showers with many muon decays _
e.g. _ , @xmath119 .
a higher ratio drives @xmath48 up and @xmath49 down , and vice versa . for our actual analysis
, we recover the flavor ratios @xmath120 of the ensemble via an unbinned maximum likelihood approach . the likelihood function for a test flavor composition @xmath121 , with @xmath122 ,
is @xmath123 the gaussian term , @xmath124 , constrains the muon component from deviating too much from its true value , assuming it can be measured from a separate track analysis .
we choose the @xmath117 width of the gaussian to be 0.12 , consistent with the present icecube measurement @xcite .
the maximum value of the likelihood determines the best - fit values of @xmath48 , @xmath49 , and @xmath50 . to estimate the uncertainty on this value , we repeat the maximum likelihood procedure using 1000 random realizations of the real data . figure [
fig : flavor_triangle ] shows the best - fit values and uncertainties on @xmath48 and @xmath49 that result from this procedure , assuming ensembles of @xmath116 showers each and real flavor ratios @xmath125 , with @xmath126 varying in the range @xmath127 $ ] .
in the main text , we consider showers of 100 tev ; the normalized distribution of number of muon decays for this shower energy is shown in fig .
[ fig : muon_decay_distribution ] .
figure [ fig : muon_decay_energies ] shows the distributions at 10 tev and 1 pev . the same general shapes and behavior of the curves is seen at all energies : @xmath0-initiated cc showers have appreciably fewer muon decays than @xmath1-initiated cc showers and nc showers .
the main change is in the intensity of the muon echo , which scales roughly linearly with shower energy . as the shower energy changes , there are moderate changes in the results .
the value of @xmath128 decreases with increasing energy , which means that @xmath0-initiated cc showers become more leptonic . and
the @xmath18 distributions for @xmath21 and @xmath22 become more similar at higher energies , and , therefore , so do their muon decay distributions .
therefore , the separation between @xmath0 and @xmath1 becomes cleaner at higher energies .
this is evidenced by contrasting the panels in fig .
[ fig : muon_decay_energies ] .
cc shower .
note the different axis scales . ] like the muon echo , the neutron echo is a product of the hadronic component of a shower .
figure [ fig : neutron_capture ] shows that the number of muon decays and the number of neutron captures is tightly correlated on an event - by - event basis . because of this , the probability distributions of the numbers of neutron captures behave similarly to those of muon decays ( figs .
[ fig : muon_decay_distribution ] and [ fig : muon_decay_energies ] ) , except for a scaling of the @xmath53-axis by a factor of about 10 .
if we were to incorporate neutron echoes in our sensitivity estimate , eq .
( [ eq : likelihood ] ) would have an extra term @xmath129 on the right - hand side , with @xmath130 the number of neutron captures in each shower of the ensemble .
however , the distribution of number of neutron captures , @xmath131 , is essentially just @xmath132 scaled up by a factor of 10 .
therefore , adding it to the likelihood would not alter the best - fit values of @xmath48 and @xmath49 or their uncertainties .
this is true from a theoretical perspective .
however , from an experimental perspective , neutron echoes are attractive because there seems to be less pmt afterpulsing at late times .
finally , there is a third possible post - shower signal the spallation echo coming from the collective cherenkov light from beta decays of long - lived ( @xmath13310 s ) unstable nuclei .
these isotopes , which are a background in low - energy neutrino detectors , are produced more efficiently in hadronic than electromagnetic showers , by a factor @xmath10 @xcite .
while the spallation echo is not observable in icecube or similar detectors due to ambient backgrounds , it might have an application in another context .
figure [ fig : flavor_triangle_flav_assump ] shows the flavor sensitivity , using muon echoes , for three different assumptions of the flavor composition at earth , including the one shown in fig . [ fig : flavor_triangle ] . for the choice of flavor composition in fig .
[ fig : flavor_triangle ] , the average @xmath117 uncertainty was 0.07 . for @xmath134 , with
@xmath135 $ ] , the best - fit values lie on the left axis of the plot ; only the one - sided @xmath117 range , of size @xmath136 , is visible . for @xmath137 ,
the best - fit values lie on the right axis of the plot ; the one - sided @xmath117 range , of size @xmath138 , is visible .
these are two extreme choices .
their smaller uncertainties are due to the fact that the total distribution of muon decays of the shower ensemble is dominated by the distribution from either @xmath0-initiated or @xmath1-initiated cc showers .
hence , our nominal choice of flavor composition , in fig .
[ fig : flavor_triangle ] , was conservative , as it has the largest uncertainty .
at fixed shower energy , the uncertainty on the @xmath0 fraction scales as @xmath139 , subject to some caveats .
when @xmath140 is small ( @xmath141 ) , the likelihood is basically flat , and one typically can not break the @xmath0-@xmath1 degeneracy with good precision .
when @xmath140 is large ( @xmath142 ) , one should take a narrower prior on the @xmath23 fraction to reflect its measurement being correspondingly better . | the flavor composition of high - energy astrophysical neutrinos is a rich observable .
however , present analyses can not effectively distinguish particle showers induced by @xmath0 versus @xmath1 .
we show that this can be accomplished by measuring the intensities of the delayed , collective light emission from muon decays and neutron captures , which are , on average , greater for @xmath1 than for @xmath0 .
this new technique would significantly improve tests of the nature of astrophysical sources and of neutrino properties .
we discuss the promising prospects for implementing it in icecube and other detectors .
* introduction. * high - energy astrophysical neutrinos , long sought , were recently discovered by the icecube collaboration @xcite .
their energy spectrum provides important clues about extreme astrophysical sources as well as neutrino properties at unexplored energies .
however , pressing mysteries remain . exploiting the flavor composition
the ratios of the fluxes of @xmath2 , @xmath3 , and @xmath4 to the total flux offers crucial additional clues . in the nominal scenario , a composition of @xmath5 at the source
is transformed by neutrino vacuum mixing to @xmath6 at earth @xcite . even for arbitrary flavor composition at the source , the maximal range of flavor composition at earth with only standard
mixing is surprisingly narrow @xcite , making deviations sensitive indicators of new physics @xcite .
so far , icecube measurements of the flavor composition mostly separate muon tracks made primarily by charged - current ( cc ) @xmath7 interactions from particle showers made by all other interactions .
a significant limitation is their poor ability to distinguish between cc interactions of @xmath0 and @xmath1 ( unless noted , @xmath8 refers to @xmath9 ) . *
synopsis of the paper. * we propose a new technique to break this @xmath0-@xmath1 degeneracy , one that could work for a wider range of energies than existing ideas ( glashow resonance @xcite , double pulses @xcite , double bangs @xcite , and lollipops @xcite ) .
we introduce two new shower observables . in showers , low - energy muons and neutrons
are produced ; after delays , the muons decay and the neutrons capture .
we call the collective cherenkov emission from the many independent decays and captures the _ muon echo _ and the _ neutron echo_. we show that the echoes are brighter for @xmath1-initiated than for @xmath0-initiated showers , which could allow them to be distinguished on a statistical basis . our focus is pointing out new observables to help solve the important problem of flavor identification .
the technical aspects of implementation require experimental expertise .
nevertheless , in a preliminary evaluation , grounded in the measured properties of icecube , we find the detection prospects promising .
figure [ fig : flavor_triangle ] shows that the present @xmath0-@xmath1 degeneracy in icecube elongates the contours of the measured flavor composition @xcite .
it also shows how detecting echoes could refine these measurements , probing the flavor composition better than the maximal range with standard mixing @xcite , which would lead to powerful conclusions . *
high - energy neutrino signatures. * at present , icecube identifies neutrino - initiated events only as tracks and showers , for which the cherenkov light appears to emanate from approximate lines and spheres .
tracks are caused by muons , which travel up to @xmath10 km in ice @xcite , due to their low interaction and decay rates .
showers are caused by all other neutrino - induced particles and extend only @xmath11 10 m in ice @xcite , due to the high interaction and decay rates of their constituent particles .
neutrinos produce secondaries through deep - inelastic scattering @xcite .
a neutrino interacts with a nucleon @xmath12 via the cc channel @xmath13 or the neutral - current ( nc ) channel @xmath14 , where @xmath15 , and @xmath16 represents hadrons .
a fraction @xmath17 of the neutrino energy goes to the final - state lepton ; the remaining fraction @xmath18 goes to the final - state hadrons .
the inelasticity distribution peaks at @xmath19 and has an average @xmath20 at 100 tev , for both @xmath21 and @xmath22 , cc and nc .
tracks are produced by @xmath23 cc interactions plus 17% of @xmath1 cc interactions where the tau decays to a muon @xcite .
showers are produced by all other neutrino interactions . for @xmath0
cc interactions , the electron- and hadron - initiated showers combine , and their sum energy equals the neutrino energy . for @xmath1
cc interactions , the tau decays promptly , so again the showers combine ( when the tau does not decay to a muon ) ; the neutrino energy estimate is slightly biased because @xmath24 of its energy is lost to outgoing neutrinos from tau decay . for nc interactions of all flavors ,
the hadron - initiated shower carries a fraction @xmath18 of the neutrino energy ; because of the steeply falling neutrino spectrum , nc interactions are subdominant in the total shower spectrum @xcite .
( this is also true for mis - identified @xmath23 cc interactions that appear to be a shower event because the track is missed @xcite . )
these points explain the basic features of the icecube results in fig .
[ fig : flavor_triangle ] .
because there are track events , the @xmath23 component of the flux must be nonzero ; because there are shower events , the sum of the @xmath0 and @xmath1 components must be nonzero .
the similarity of @xmath0- and @xmath1-initiated events makes the contours nearly horizontal ; the degeneracy is weakly broken because increasing the @xmath25 fraction increases the number of tracks and decreases the shower energies .
with present methods , improvement requires much larger exposure @xcite .
* electromagnetic versus hadronic showers. * the key to our new method is understanding the low - energy physics underlying high - energy showers @xcite . when showers are developing , particles multiply in number while decreasing in energy .
an electromagnetic shower starts out with electrons , positrons , and gamma rays and stays composed predominantly of them ; there is usually a small fraction of pions and nucleons produced by photonuclear processes .
a hadronic shower starts out with pions and nucleons , and then builds up a progressively larger fraction of electromagnetic particles as prompt @xmath26 decays deplete @xmath27 of the remaining hadronic energy with each shower generation .
shower development ends when the average particle energy is low enough that the particle- and energy - loss rates exceed the particle - production rates . at that point ,
the most abundant particles in all showers are @xmath28-mev electrons and positrons , which produce most of the prompt cherenkov light .
pions carry only @xmath10% of the energy in hadronic showers and @xmath29% in electromagnetic showers . however , they are the key to separating electromagnetic and hadronic showers . * new shower observables. * at the end of shower development , charged pions come to rest by ionization ; then @xmath30 capture on nuclei and @xmath31 decay to @xmath32 . the @xmath32 decay with a lifetime of 2.2 @xmath33s , producing @xmath34 with @xmath11 35 mev .
the collective cherenkov light from these positrons is our first new observable : the _ muon echo_. separately , neutrons lose energy by collisions until they reach thermal energy .
they eventually capture on hydrogen , with a timescale of @xmath35 200 @xmath33s , producing 2.2 mev gamma rays .
( in seawater , 33% of neutrons capture on cl ; the emitted gamma rays have @xmath36 mev @xcite , making the neutron echoes more visible . ) the gamma rays compton - scatter electrons to moderate energies , producing cherenkov light .
this collective emission is our second new observable : the _ neutron echo_. 10 times smaller . ]
we simulate showers and subsequent echoes using the ` fluka ` monte carlo software ( version 2011.2c-4 ) @xcite .
we inject high - energy electrons or positrons to simulate electromagnetic showers and charged pions to simulate hadronic showers .
figure [ fig : time_scale ] shows the averaged time profile of a 100-tev hadronic shower .
because the features happen on very different timescales , it is appropriate to analyze their light yield @xmath37 in bins of log time .
accordingly , we plot @xmath38 ; this makes the height of the curve proportional to its contribution to the integrated light yield .
the echo shapes are exponentials with the respective timescales .
the echoes are well - separated from the prompt shower and from each other .
figure [ fig : time_scale ] also shows that the echoes have low intensities : the muon echo has @xmath39 of the prompt shower energy and the neutron echo has @xmath40 .
the first number results from the facts that 10% of hadronic shower energy goes to pions , 10% of those pions are @xmath31 that come to rest and decay , and 30% of the pion decay energy goes to positrons from muon decays .
the second number results from the facts that there are about 10 times more neutron captures than muon decays , that the capture energy is about 20 times smaller , and that the cherenkov efficiency is about 3 times smaller . intrinsic fluctuations . ]
the points above carry over for electromagnetic showers , except for a crucial difference : the pions carry only @xmath11 1% of the shower energy as opposed to @xmath11 10% .
thus the echo intensities are expected to be @xmath11 10 times higher in hadronic showers than in electromagnetic showers .
figure [ fig : n_mu ] shows that there are indeed about 10 times as many muon decays and neutron captures in hadronic showers .
this difference is much larger than the intrinsic fluctuations of these numbers . because the number of decays and captures , and , therefore , the light coming from them , grows linearly with shower energy , this factor - of-10 difference between electromagnetic and hadronic showers is present at all energies .
the yields may have an overall shift of up to a factor of 2 due to hadronic and nuclear uncertainties @xcite , but this can be calibrated by external measurements @xcite or _ in situ_. * separating @xmath0 and @xmath1. * we now examine how echoes can be used to help identify the flavors of neutrino - induced showers . in realistic neutrino interactions ,
the differences in the echoes are less stark than above .
showers initiated by @xmath0 are mostly electromagnetic because the outgoing electron typically carries more energy than the final - state hadrons . but showers initiated by @xmath1 are mostly hadronic because , in addition to the shower from the final - state hadrons , 67% of tau decays are hadronic .
( nc showers are purely hadronic . )
we consider flavor separation at fixed shower energy , as opposed to fixed neutrino energy , to make contact with experiment .
we simulate neutrino interactions with appropriate energies to give @xmath41 tev , including 10% energy resolution @xcite . for nc interactions ,
we mimic the final - state hadrons by directly injecting charged pions at the shower energy .
figure [ fig : muon_decay_distribution ] shows how the numbers of muon decays per shower are distributed for different neutrino interaction channels .
as expected , @xmath0 cc showers produce fewer muons than @xmath1 cc showers .
the basics of the distributions in fig .
[ fig : muon_decay_distribution ] can be understood easily . for pure electromagnetic showers ,
the peak would be at @xmath42 decays ; it would be narrow because most pions are produced late in the shower and the fluctuations are mostly poissonian . for pure hadronic showers , the peak would be at @xmath43 decays ; it would be broad because there are large fluctuations in how much energy goes into @xmath44 in the first few shower generations .
the shapes shown in fig . [ fig : muon_decay_distribution ] depend also on the @xmath18 distributions for neutrino interactions . for @xmath0 cc events , the distribution is substantially broadened because the differential cross section @xmath45 , while peaked at @xmath46 , has a substantial tail . for @xmath1 cc events , there is a slight shift to the left , due to the 17% of tau decays to muons . the results in fig .
[ fig : muon_decay_distribution ] make it possible to distinguish @xmath0 and @xmath1 on a statistical basis .
we next estimate the sensitivity to flavor composition using the echoes from an ensemble of events , assuming perfect detection efficiency .
first , we use the results in fig .
[ fig : muon_decay_distribution ] to generate the muon decay distributions for each flavor , assuming an equal flux of @xmath8 and @xmath47 , and nc to cc event ratios consistent with a power - law spectral index of 2.5 @xcite .
next , for an assumed flavor composition , we randomly sample the number of muon decays for each shower in an ensemble of 100 showers of @xmath41 tev .
then , we treat the flavor composition @xmath48 and @xmath49 as free parameters ( @xmath50 ) and use an unbinned maximum - likelihood procedure to find their best - fit values .
we generate @xmath51 different realizations of the shower ensemble , and find the average best - fit values and uncertainties of @xmath48 and @xmath49 .
further details are in appendices [ sec : formalism ] and [ sec : sensitivity ] .
figure [ fig : flavor_triangle ] shows the predicted sensitivity on @xmath48 and @xmath49 , assuming equal @xmath0 and @xmath1 content , _
i.e. _ , a composition of the form @xmath52 , where @xmath53 varies in @xmath54 $ ] . the vertical shape of the band shows that the sensitivity to @xmath48 and @xmath49 does not depend on the @xmath23 content .
( because our method is only weakly sensitive to @xmath55 , we suppress its uncertainty in the plot . )
our results are conservative .
the sensitivity improves slightly with shower energy ; see appendix [ sec : muon_other_energies ]
. assuming perfect detection efficiency , the sensitivity is comparable whether we use muon echoes only , neutron echoes only , or both ; see appendix [ sec : neutron ] .
it is also comparable , or better , for other choices of input parameters ; see appendix [ sec : other_assumptions ] .
* observability of the echoes. * echo detection depends on how the echo light yield compares to that from ambient backgrounds and detector transients .
these quantities are detector - dependent , and we use icecube as a concrete example .
the echoes are faint , but they are well localized , which enhances their visibility . in space , like the parent shower , they are concentrated among only the few photomultiplier tubes ( pmt ) on a single string that are closest to the neutrino interaction vertex @xcite . in time , they occur @xmath56 @xmath33s and @xmath57 @xmath33s after the prompt shower .
these timescales require long - time data collection , made possible by the recent development of the hitspooling technique , which can go to hours for infrequent events @xcite . in direction ,
the shower light is beamed forward but the echo light is isotropic .
light scattering makes the shower more isotropic and increases its duration @xcite , which could partially obscure the muon echo .
the total light yield of a shower in icecube is @xmath28 detected photoelectrons ( p.e . ) per tev @xcite . for 100 tev ,
the muon echo in a hadronic shower is expected to yield @xmath58 p.e . and the neutron echo @xmath59 p.e . the low p.e
. counts set the energy threshold for our method .
ambient backgrounds in icecube do not eclipse the echoes . for an average p.e .
noise rate of @xmath42 hz per pmt @xcite , the expected backgrounds in @xmath60s and @xmath61s are only @xmath62 and @xmath63 p.e . per pmt , respectively .
( even with correlated noise , due to nuclear decays near the pmt , the backgrounds will be small in all but a few pmts , and those will be identifiable @xcite . ) and the cosmic - ray muon rate in icecube is 3 khz @xcite , so the probability of a muon lighting up several specific pmts in the short time between shower and echo is small .
a serious concern is the detector transient called afterpulsing , where a pmt registers late p.e . with total charge proportional to the initial signal ( the shower ) and with a time profile characteristic to the pmt . for the icecube pmts
, the muon echo will compete with an afterpulse feature of relative amplitude @xmath64 near 2 @xmath33s @xcite ; though larger than the echo , it is not overwhelmingly so .
encouragingly , the neutron echo , though smaller , is late enough that afterpulsing seems to be negligible . in summary
, the prospects for observing echoes are promising , and they improve with shower energy .
doing so may require changes in detector design or in pmt technology @xcite ; these considerations may shape the design of icecube - gen2 @xcite , km3net @xcite , and baikal - gvd @xcite .
with multiple nearby pmts @xcite , it may be possible to reconstruct individual events , dramatically improving background rejection . the final word on
the observability of the echoes will come from detailed studies by the experimental collaborations . * conclusions. * the rich phenomenology contained in the flavor composition of high - energy astrophysical neutrinos can not be fully explored due to the difficulty of distinguishing showers initiated by @xmath0 versus @xmath1 in neutrino telescopes . to break this degeneracy ,
we have introduced two new observables of showers : the delayed , collective light , or `` echoes , '' from muon decays and neutron captures .
this light reflects the size of the hadronic component of a shower , and it is stronger in @xmath1-initiated than @xmath0-initiated showers . figure [ fig : flavor_triangle ] shows the promise of our method for icecube . with assumptions of 100 showers and perfect detection efficiency , echo measurements would improve the separation of @xmath0 and @xmath1 by a factor of @xmath65 over present measurements .
that is comparable to the estimated sensitivity attainable with the present technique after more than 50 years of exposure of the next - generation detector icecube - gen2 , assuming it will have an effective area 6 times larger than icecube .
the applications of tagging hadronic showers via muon and neutron echoes extend beyond flavor discrimination .
the technique could improve shower energy reconstruction , by folding in the probability of a shower being electromagnetic or hadronic . and , at the considered energies ,
the echoes are shifted forward along the shower direction by @xmath66 m from the shower peak .
if this shift can be detected , it would improve the poor angular resolution of showers @xcite .
high - energy neutrino astronomy has just begun .
we are still learning the best ways to detect and analyze astrophysical neutrinos .
we should pursue all potentially detectable signatures , edging closer to finding the origins and properties of these ghostly messengers .
we thank carlos argelles , amy connolly , eric huff , tim linden , kenny ng , david nygren , eric obela , annika peter , benedikt riedel , jakob van santen , shigeru yoshida , guanying zhu , and especially lutz kpke and anna steuer for useful discussions and comments .
we also thank markus ahlers , kfir blum , claudio kopper , carsten rott , and others for helpful feedback beginning at jfb s presentation of these ideas at the ccapp cosmic messages in ghostly bottles workshop in 2014 .
swl , mb , and jfb are supported by nsf grant phy-1404311 .
76ifxundefined [ 1 ] ifx#1 ifnum [ 1 ] # 1firstoftwo secondoftwo ifx [ 1 ] # 1firstoftwo secondoftwo `` `` # 1''''@noop [ 0]secondoftwosanitize@url [ 0 ]
+ 12$12 & 12#1212_12%12@startlink[1]@endlink[0]@bib@innerbibempty link:\doibase 10.1103/physrevlett.111.021103 [ * * , ( ) ] , link:\doibase 10.1126/science.1242856 [ * * , ( ) ] , link:\doibase 10.1103/physrevd.89.062007 [ * * , ( ) ] , link:\doibase 10.1103/physrevlett.113.101101 [ * * , ( ) ] , link:\doibase 10.1088/0004 - 637x/809/1/98 [ * * , ( ) ] , link:\doibase 10.1103/physrevlett.115.081102 [ * * , ( ) ] , link:\doibase 10.1016/0927 - 6505(94)00043 - 3 [ * * , ( ) ] , link:\doibase 10.1103/physrevd.62.103007 [ * * , ( ) ] , link:\doibase 10.1103/physrevlett.115.161302 [ * * , ( ) ] , link:\doibase 10.1103/physrevlett.90.181301 [ * * , ( ) ] , link:\doibase 10.1103/physrevd.67.073024 [ * * , ( ) ] , link:\doibase 10.1103/physrevd.68.093005 [ * * , ( ) ] , , link:\doibase 10.1103/physrevd.69.017303 [ * * , ( ) ] , link:\doibase 10.1103/physrevlett.92.011101 [ * * , ( ) ] , link:\doibase 10.1103/physrevd.74.013010 [ * * , ( ) ] , link:\doibase 10.1103/physrevd.75.123005 [ * * , ( ) ] , link:\doibase 10.1088/1126 - 6708/2008/02/005 [ * * , ( ) ] , link:\doibase 10.1016/j.nuclphysb.2009.06.017 [ * * , ( ) ] , link:\doibase 10.1103/physrevd.80.103005 [ * * , ( ) ] , link:\doibase 10.1142/s0217751x09047429 [ * * , ( ) ] , link:\doibase 10.1103/physrevd.80.113006 [ * * , ( ) ] , link:\doibase 10.1007/jhep04(2010)066 [ * * , ( ) ] , link:\doibase 10.1007/jhep05(2011)133 [ * * , ( ) ] , link:\doibase 10.1088/1475 - 7516/2012/10/020 [ * * , ( ) ] , link:\doibase 10.1103/physrevlett.113.091103 [ * * , ( ) ] , link:\doibase 10.1103/physrevd.90.065035 , 10.1103/physrevd.90.089904 [ * * , ( ) ] , link:\doibase 10.1088/1475 - 7516/2014/12/039 [ * * , ( ) ] , link:\doibase 10.1103/physrevd.91.053001 [ * * , ( ) ] , link:\doibase 10.1103/physrevd.91.103008 [ * * , ( ) ] , link:\doibase 10.1103/physrevlett.114.171101 [ * * , ( ) ] , link:\doibase 10.1103/physrevlett.114.171102 [ * * , ( ) ] , link:\doibase 10.1140/epjc / s10052 - 015 - 3664 - 6 [ * * , ( ) ] , link:\doibase 10.1103/physrevlett.115.161303 [ * * , ( ) ] , link:\doibase 10.1140/epjc / s10052 - 016 - 3893 - 3 [ * * , ( ) ] , link:\doibase 10.1103/physrevd.93.085004 [ * * , ( ) ] , @noop ( ) , link:\doibase 10.1103/physrev.118.316 [ * * , ( ) ] link:\doibase 10.1016/j.physletb.2005.06.056 [ * * , ( ) ] , link:\doibase 10.1088/1475 - 7516/2011/10/017 [ * * , ( ) ] , link:\doibase 10.1103/physrevd.93.022001 [ * * , ( ) ] , link:\doibase 10.1016/j.astropartphys.2006.11.003 [ * * , ( ) ] , link:\doibase 10.1016/0927 - 6505(96)00008 - 4 [ * * , ( ) ] , link:\doibase 10.1103/physrevd.58.093009 [ * * , ( ) ] , link:\doibase 10.1103/physrevd.83.113009 [ * * , ( ) ] , link:\doibase 10.1088/1674 - 1137/38/9/090001 [ * * , ( ) ] link:\doibase 10.1088/1475 - 7516/2004/11/009 [ * * , ( ) ] , @noop _ _ , ed .
( , ) link:\doibase 10.1016/j.astropartphys.2004.09.003 [ * * , ( ) ] link:\doibase 10.1103/physrevd.79.063001 [ * * , ( ) ] , link:\doibase 10.1016/j.nuclphysbps.2009.09.060 [ * * , ( ) ] link:\doibase 10.1103/physrevd.88.055005 [ * * , ( ) ] , link:\doibase 10.1103/physrevc.89.045801 [ * * , ( ) ] , link:\doibase 10.1103/physrevd.91.105005 [ * * , ( ) ] , link:\doibase 10.1103/physrevd.92.105033 [ * * , ( ) ] , @noop `` , '' , @noop ( ) link:\doibase 10.1063/1.2720455 [ * * , ( ) ] link:\doibase 10.1103/physrevc.81.025807 [ * * , ( ) ] , link:\doibase 10.1088/1475 - 7516/2013/08/049 [ * * , ( ) ] , link:\doibase 10.1103/physrevd.93.012004 [ * * , ( ) ] , link:\doibase 10.1103/physrevlett.93.171101 [ * * , ( ) ] , @noop ( ) , @noop ( ) , link:\doibase 10.1016/j.nuclphysbps.2015.09.050 [ * * , ( ) ] link:\doibase 10.1088/1748 - 0221/9/03/p03009 [ * * , ( ) ] , @noop `` , '' , @noop ( ) , link:\doibase 10.1016/j.nima.2010.03.102 [ * * , ( ) ] , _ _ , @noop master s thesis , ( ) link:\doibase 10.1364/ao.41.004975 [ * * , ( ) ] @noop ( ) , link:\doibase 10.1016/j.nima.2006.05.235 [ * * , ( ) ] , @noop ( ) , link:\doibase 10.1016/j.nima.2012.11.151 [ * * , ( ) ] @noop `` , '' , link:\doibase 10.1103/physrevd.88.043009 [ * * , ( ) ] , * supplemental material * in the main text , we showed how detecting muon echoes can improve discrimination between @xmath0-initiated and @xmath1-initiated showers .
we showed results using 100-tev showers and a flavor composition of the form @xmath67 . here
we provide more details on the statistical method and show how the results depend on choices of inputs . in appendix
[ sec : formalism ] , we present the underlying formalism for flavor discrimination per shower . in appendix
[ sec : sensitivity ] , we apply it to an ensemble of showers . in appendix
[ sec : muon_other_energies ] , we discuss flavor discrimination at other shower energies . in appendix
[ sec : neutron ] , we discuss neutron echoes . in appendix
[ sec : other_assumptions ] , we show sensitivity results for other input choices . to simplify the notation , we explicitly show the shower energy @xmath68 dependence when defining a quantity , and suppress it otherwise . in the probability definitions ,
we show @xmath21 cc cases explicitly ; nc and @xmath69 cases have similar definitions , with cc replaced by nc , and @xmath21 replaced by @xmath22 . |
knowing the spectrum of quantum mechanical states of an electron in magnetic field there are two ways of calculating the thermal gyroradius at given temperature @xmath24 .
either the gyroradius is calculated by averaging it over the distribution which is its expectation value @xmath13 , would be the exact way to do it , or one calculates the energy expectation value @xmath32 which provides the expectation value of the squared gyroradius @xmath33 whose root can also be taken to represent the thermal gyroradius .
this latter value will be calculated first .
the energy levels of an electron in a homogeneous magnetic field ( the case of interest here as a magnetic inhomogeneity provides only higher order corrections ) has been calculated long ago @xcite .
since the parallel dynamics of an electron is not of interest here , it can be taken classically .
then the energy of an electron in the magnetic field becomes @xmath34 with @xmath35 , quantum number @xmath36 , and @xmath37 the two spin directions .
the average distribution of electrons over these @xmath36 perpendicular energy states is given by the fermi distribution @xmath38 written here for the anisotropic case , with @xmath39 the fugacity which depends on the chemical potentials @xmath40 in parallel and perpendicular direction , and @xmath41 are the inverse temperatures ( in energy units ) .
at sufficiently large temperatures the unity in the denominator is neglected , a very well justified approximation which still takes into account the non - continuous energy distribution over discrete landau levels thus maintaining the quantum character of electrons .
the fugacities enter into the normalization constant now .
this is the case far away from eq .
( [ denslim ] ) which for plasmas interests us here . under these conditions
the expectation value of the ( average ) perpendicular energy of the electrons ( i.e.the perpendicular electron pressure ) is calculated from the integral @xmath42 the spin contribution in the perpendicular energy either compensates for the half landau energy level or completes it to the first order level .
thus the sum splits into two terms which both are geometric progressions which can immediately be done .
the final result , using the normalization of the integral to the average density of particles and dividing through @xmath43 thus yields for the average energy @xmath44 at the assumed large temperatures the exponentials must be expanded to first order yielding the very well known and expected classical result that the average energy is the temperature , @xmath45 .
hence , taking its root and inserting into the gyroradius we find what is expected in this case : @xmath46 this is the root - mean - square gyroradius , a well known quantity . at lower temperatures @xmath47 , still by far exceeding fermi energy .
the former expression for @xmath32 has to be used in this expression . however , the correct gyroradius is not its root mean square but the expectation value @xmath48 .
this is substantially more difficult to calculate .
there are two ways of finding the expectation value .
either one makes use of the landau solution of the electron wave function and refers to the wigner distribution of quantum mechanical states .
this , for our purposes would be overdoing the problem substantially .
we do not need the quantum mechanical probability distribution for the simple estimate we envisage here .
afterwards we would anyway have to combine the wigner function with a suitable momentum distribution .
the second and simpler way is to refer to the above fermi distribution and directly using the energy distribution as above . under the same conditions as in the calculation of the rms value
, this procedure circumvents the use of the wave function being interested only in the energy levels .
it , however , requires the solution of the integral - sum @xmath49 the sum , the integral contains , can not anymore be done in closed form as there is no known way to tackle the summation over the root quantum index @xmath36 in a non - geometric progression .
( we may note in passing that for a non - moving plasma calculating the average velocity moment would lead to a null result .
however , in calculating the expectation value of the gyroradius , the perpendicular velocity , being a radius vector in perpendicular velocity space , is always positive .
this is reflected by the summation over positive quantum numbers @xmath36 only .
it and the gyroradius are positive definite quantities which both do not average out . )
one thus needs to deviate from summing and to approximate the sum by an integral , which slightly overestimates the final result . transforming the energy into a continuous variable , which implies that the summation index becomes continuous , then simply leaves us with a gaussian integral of which we know that the mean value and the rms values are related by the classical formula @xmath50 the classical result for the mean , where by the @xmath51 sign we indicated that the integral yields an upper limit for the expectation value of the gyroradius .
the above estimates permit determining the thermal fluctuation ( thermal spread ) of the gyroradius .
this fluctuation is the difference between the mean square and squared expectation values . with the above results we find that the thermal fluctuation of the gyroradius amounts to @xmath52 this number falls into the @xmath0 range of `` freedom '' in choosing the thermal speed that is commonly used in the definition of the rms gyroradius as either @xmath53 or @xmath54 .
the expected result of the quantum mechanical calculation of the electron gyroradius of a hot plasma is that the root - mean - square value of the gyroradius is reproduced by quantum mechanics in the high temperature limit which clearly completely justifies its use not adding anything new except for the confirmation of well - known facts which can be generated in much simpler way by using the boltzmann calculation . for lower temperatures we obtained a slightly more precise expression for the average energy which should be used in the definition of the thermal gyroradius .
the expectation value of the gyroradius is different from the root mean square value , which is also known .
its value is lower given by the gaussian mean .
usually this value is not used in any classical calculation .
the difference between both values is within the commonly used @xmath0 freedom for choosing the rms thermal speed .
the above calculation is based on keeping the discrete distribution of all plasma electrons over landau levels .
since these are equally spaced up to infinity , the energy remains discontinuous even at large temperatures .
the correction is , however , unimportant .
it is sometimes claimed that the discrete distribution of electron energies is reflected in the dielectric function of a plasma which leads to collective cyclotron harmonics , i.e. bernstein modes .
however , landau levels have nothing in common with cyclotron harmonics .
they are single particle energy levels occupied at most by two electrons of oppositely directed spins , while cyclotron harmonics are eigenmodes of the magnetized plasma , channels which allow for propagation of electro(magnetic ) signals .
the discrete energy distribution of the electrons naturally disappears in the cacluation of the expectation values and is not reflected in any cyclotron harmonic waves .
the average gyroradius appearing in the arguments of the modified bessel functions in the dielectric expression of the harmonics in the combination @xmath55 is a
_ mere scaling factor _ of wavelengths related to the perpendicular thermal speed .
it can not be taken as definition of a gyroradius , which is a particle property .
in fact , calculation of the dielectric function from quantum mechanics accounting for the landau energy distribution is very difficult . in classical theory one integrates the vlasov equation with respect to time along classical particle orbits @xmath56 which , in quantum mechanics , is forbidden as _ no orbits exist_. orbits are replaced by probability distributions in space varying with time .
the calculation therefore needs to be performed via a feynman path integral approach which has not yet been done .
it is , however , reasonably expected that it will , in the classical limit reproduce the classical dielectric function and the harmonic structure of bernstein and cyclotron modes .
sonnerup , b. u. . : magnetic field reconnection , in : solar system plasma physics , vol .
45 - 108 , edited by : lanzerotti , l. t. , kennel , c. f. , and parker , e. n. , north - holland publ .
, amsterdam - new york , 1979 . | the average ( thermal ) gyroradius of charged particles is re - examined from a quantum - mechanical point of view .
the straight quantum - mechanical calculation clearly reproduces its conventionally used random - mean - square ( rms ) expectation value .
the quasi - classical approach reproduces its mean expectation value as well thus confirming the purely classical calculation .
it shows that the fluctuations in the gyroradius amount to 21.5% of its rms value .
this fluctuation is , however , within the usual @xmath0-``freedom '' range of choice in the definition of the rms gyroradius respectively the mean thermal speed on which the rms value is based , its correct ( nonrelativistic ) value @xmath1 and its simplified version @xmath2 .
the gyroradius @xmath3 of a particle of charge @xmath4 and mass @xmath5 in a magnetic field of strength @xmath6 is one of the fundamental parameters used in plasma physics .
its exact classical definition for a single particle of perpendicular velocity @xmath7 is @xmath8 written here for an electron with cyclotron frequency @xmath9 ( see any textbook on plasma physics , e.g. , * ? ? ?
* ) .
its ion equivalent is obvious .
single electrons have undefined temperature , usually assumed to be zero , hence the exactness of the given expression . in quantum mechanics electrons obey landau levels possessing a well defined magnetic length @xmath10 which corresponds to the gyroradius of an electron in the lowest landau energy level , @xmath11 .
in high temperature plasmas one defines the gyroradius through the thermal velocity , respectively the perpendicular temperature @xmath12 .
a more precise definition makes use of the velocity distribution function of the electrons to calculate the average gyroradius @xmath13 as the moment of the classical distribution with respect to the perpendicular particle velocity . in the random - mean - square ( rms ) limit this yields the above expression for @xmath14 with replacement @xmath15 , which is nothing but the well - known rms thermal gyroradius " .
occasionally freedom is taken for convenience in either keeping or dropping the @xmath0 value in the definition of the thermal speed which introduces a 29% indeterminacy of the thermal speed and rms gyroradius . for temperatures @xmath16 , i.e. @xmath17 , with @xmath18 fermi temperature ,
electrons become degenerate and one needs to refer to the fermi distribution . in this case the thermal speed is replaced by the fermi velocity @xmath19 , a _ collective _ velocity of degenerate electrons .
one may note that this does _ not _ reproduce the above quantum mechanical magnetic length of electrons , @xmath20 ! the latter being a single electron property in magnetic fields caused by the quantum character of the magnetic flux with flux quantum @xmath21 , not a collective effect .
it defines the radius of a magnetic field line through @xmath22 . at non - zero temperatures @xmath23 slightly exceeding @xmath18
a small correction is to be introduced on the fermi velocity entering the collective rms gyroradius . in many particle theory ,
starting from the single particle picture one should , for greater precision , make use however of the distribution of particles over the entire spectrum of energy states available and accessible to the population at a given temperature @xmath24 .
since only two particles of opposite spins can occupy any energy state there is a well - known competition between the available states and the number of particles at @xmath24 . if the temperature @xmath25 , with @xmath26 the fermi temperature , then the particles have effectively zero temperature , and the gyroradius is defined through their single velocity in the respective state .
this case is encountered in high density compounds at @xmath27}}^\frac{3}{2 } \quad \mathrm{m}^{-3}\ ] ] with temperature measured in ev .
( one may note that this number is remarkably close to the conditions encountered in the solar chromosphere ! ) .
when this condition holds , quantum effects become important ; it implies that the interparticle distance @xmath28 equals the thermal de broglie wavelength @xmath29 , here written with @xmath30 . at higher temperatures and lower densities the average gyroradius should be calculated by adding up all electrons in the available states . by assuming a classical distribution function this is circumvented and reduced to a simple integration yielding the above thermal rms value @xmath31 . here
we present the exact quantum mechanical calculation which naturally confirms the thermal rms gyroradius result but , in addition , yields the expectation value @xmath13 of the gyroradius and provides its width of uncertainty , i.e its thermal spread . |
the crab nebula has been the best laboratory for investigating the mechanism linking the pulsar wind nebula ( pwn ) with the pulsar .
the energetics confirms that the pwn is powered by the spin - down energy of the pulsar , and it is generally believed that this energy is transported by a relativistic wind ( e.g. kennel & coroniti 1984 ) .
hester ( 1998 ) and tanvir , thomson , & tsikarishvili ( 1997 ) showed that the `` wisps '' , which are elliptical ripples around the pulsar ( scargle 1969 ) , are moving outwards with a speed of about 0.5_c_. such high energy phenomena must be associated with x - ray emission .
here , we present the results of a series of monitoring x - ray observations of the crab nebula with _
chandra _ , whose spatial resolution of @xmath0 is comparable to that of ground - based optical telescopes .
we adopt a distance of 2 kpc to the crab nebula throughout this paper .
the crab nebula was observed with acis - s3 ( the back - illuminated ccd chip ) eight times , once every three weeks from 2000 november 3 to 2001 april 6 .
these observations were coordinated with the _ hubble space telescope _ ( hst ) ( hester et al . 2002 ) .
each observation has approximately 2.6 ksec of effective exposure time .
we employed a 0.2 sec frame time to reduce pileup and a restricted window to reduce dead - time .
the window size is only slightly larger than the x - ray extent of the crab nebula ( see fig [ fig:2ndimage ] ) .
all of the images shown were made using the 0.210 kev band .
figure [ fig:2ndimage ] shows one of eight images of the crab nebula .
it clearly shows an axisymmetric structure about the polar jet with the torus and the inner ring resolved in an early _ chandra _ observation ( weisskopf et al .
fine fibrous structures are also resolved at periphery of such large - scale structures .
they show clear correlation with a optical polarization measurement ( hickson & van den bergh 1990 ) , indicating that they trace local magnetic field structures .
figure [ fig:2468image ] shows a series of the observations .
two wisps moving outward are detected through all of eight observations .
they can also be seen in simultaneous hst observations ( hester et al .
2002 ) . here ,
we denote them as `` wisp a '' and `` wisp b '' .
they appear to break off from the inner ring .
with respect to the inner ring s elliptical shape , the shapes of the wisps look warped due to the time delay of light travel ( hester et al . 2002 ) .
we measured the speed of the wisps , assuming that they are moving in the equatorial plane which includes the inner ring , and taking the inclination of the plane ( the inclination angle was derived assuming the inner ring is circular ) and the time delays into account . in spite of the difference in the directions and the birth times , the speeds of the two wisps are almost same ,
@xmath10.43_c_. their similarity indicates the existence of the continuous isotropic pulsar wind in the equatorial plane .
the inner ring is also variable , but unlike the wisps , it preserves its overall ring - like shape and relative position with respect to the pulsar .
the ring never forms a continuous loop , but appears intermittent and mostly consists of knot - like features . among them , three knots , which lie along the southeast portion of the ring and are symmetric about the axis of the jet , gradually brighten by factor @xmath11.5 within our observational period of 6 months .
however , we note that these knots were bright enough to be detected 1 year before our 1st observation ( weisskopf et al .
additionally , some blob - like features appear to move outward along the jet , with a speed of @xmath10.34_c_. temporal variations can also be seen in the torus .
figure [ fig : diff ] shows the difference images of the 2nd@xmath23rd and the 2nd@xmath28th observations .
although the variation is not strongly pronounced in the 3-week difference image , except for the wisps , differences within the torus and along its boundary are quite substantial over a duration of 19 weeks .
the torus seems to be expanding at 0.10.2_c_. due to the small displacement and the ambiguous boundary , relatively large uncertainties remain .
the fact that the angular extent along the major axis measured with _
chandra _ agrees well with those measured with _
( brinkmann et al . 1985 )
, _ rosat _ ( hester et al . 1995 ; greiveldinger & aschenbach 1999 ) , and even from lunar occultation 25 years ago ( aschenbach & brinkmann 1975 ) suggests that the torus is stable on tens of arcsecond scales over decades , but varies on arcsecond scales on several months .
the northwestern region and the end of the jet do not exhibit any strong variations .
these x - ray observational pictures generally match the canonical scenario that the pwn is confined by the supernova ejecta .
the relativistic winds from the pulsar are continuously transported in the equatorial plane and accumulate at the torus , outside of which the optical filaments ejecta can be seen .
aschenbach , b. , & brinkmann , w. 1975 , , 41 , 147 brinkmann , w. , aschenbach , b. , & langmeier , a. 1985 , nature , 313 , 662 greiveldinger , c. , & aschenbach , b. 1999 , , 510 , 305 hickson , p. & van den bergh , s. 1990 , , 365 , 224 hester , j. j. et al .
1995 , , 448 , 240 hester , j. j. 1998 , in neutron stars and pulsars : thirty years after the discovery , ed .
n. shibazaki et al . , 431 hester , j. j. , et al .
2002 , in neutron stars in supernova remnants , ed .
slane & b.m .
gaensler , in press .
kennel , c. f. , & coroniti , f. v. 1984 , , 283 , 694 pavlov , g. g. , kargaltsev , o. y. , sanwal , d. , & garmire , g. p. 2001 , 554 , 189 scargle , j. d. 1696 , , 156 , 401 tanvir , n. r. , thomson , r. c. , & tsikarishvili , e. g. 1997 , new astronomy , 1 , 311 weisskopf , m. , et al .
2000 , , 536 , 81 | we present a series of monitoring observations of the crab nebula with the _ chandra x - ray observatory _ , focusing on the temporal evolution of the structure .
this series of 8 observations , spanning a period of approximately six months , shows the dynamic nature of the inner x - ray structures .
we detected outward moving `` wisps '' from the recently discovered inner ring seen in optical observations .
we also find that the inner ring itself shows temporal variations in structure .
the torus also appears to be expanding .
such temporal variations generally match the canonical scenario that an expanding synchrotron nebula injected from the pulsar is confined by the supernova ejecta .
# 1_#1 _ # 1_#1 _ = # 1 1.25 in .125 in .25 in |
classical novae are cataclysmic variable stars which are semidetached binary systems containing a late - type star and a white dwarf ( wd ) .
the nova outburst is induced by a thermonuclear runaway in the matter transferred from the late - type star onto the surface of the wd .
a large amount of gas accumulated on the wd is ejected due to nova outbursts .
the ejected mass is typically @xmath7 m@xmath8 in an outburst .
the expansion velocity of the ejecta reaches @xmath9@xmath10 .
the outburst amplitude of novae is @xmath11@xmath12 mag in optical @xcite .
dust formation episodes are occasionally observed during nova outburst phases .
the dust formation occurs 30 80 d after the maximum of the light curve .
this phase is the `` transition phase '' @xcite .
we can study the formation and cooling processes of dust through the temporal evolution of novae .
v2362 cyg (= nova cygni 2006 ) was discovered on april 2.807 ut by h. nishimura at @xmath13 @xcite .
@xcite spectroscopically confirmed that it was a classical nova just around its maximum from the h@xmath14 emission line with a clear p - cygni profile .
@xcite reported that the nova belongs to the fe class .
@xcite identified the progenitor of v2362 cyg on their galactic plane survey images taken on 2004 aug 3 as a point source with the magnitudes of @xmath15 and @xmath16 , which indicates the outburst amplitude of @xmath17 .
the nova experienced an unusual rebrightening event from @xmath18 d after the first maximum .
the nova reached its rebrightening maximum on 2006 december 1.9 (= jd 2454071.4 ; @xcite ) .
a similar rebrightening event has been found only in v1493 aql @xcite .
v2362 cyg is the second case which exhibited such a rebrightening , and hence , has received much attention @xcite .
most recently , v2491 cyg showed a small rebrightening event and some authors suggested that v2491 cyg belongs to the same class @xcite .
such rebrightening events are totally unexpected from the standard picture of classical novae , and their mechanism is still unclear . optical near - infrared multiband monitoring would be crucial to explore the temporal evolution of the photosphere and outflow material around the rebrightening event .
we performed extensive optical and near - infrared ( nir ) photometric observations of v2362 cyg covering the early decline and the rebrightening event of the object .
our data trace the evolution of outflow materials around the rebrightening event . in section 2
, we describe our observations .
the results are presented in section 3 .
we discuss the characteristics of the dust formation in section 4 . finally , we summarize our findings in section 5 .
we performed simultaneous optical and nir observations during the rebrightening event with trispec attached to the kanata 1.5-m telescope at higashi - hiroshima observatory .
trispec is an imager and spectrograph with a capability of polarimetry covering both optical and nir wavelengths ( @xcite ) .
the trispec observation was conducted from 2006 november 20.56 ut ( jd 2454060.06 ) , 11 d before the rebrightening maximum , to 2007 february 5.87 ( jd 2454137.37 ) , 68 d after past the rebrightening maximum .
we obtained typically 10 frames in a night with different dithering positions in individual bands .
exposure times in the @xmath19 , @xmath20 , @xmath21 , @xmath22 , @xmath23 , @xmath24 and @xmath25 in a night were typically 450 , 380 , 150 , 155 , 140 , 180 and 75 s , respectively .
we have no @xmath23- and @xmath24-band images between 2006 december 15 and 2007 january 18 because of a hardware problem .
we obtained differential magnitudes of the object relative to a comparison star using aperture photometry with the apphot package in iraf .
we used a nearby star tyc 3181 - 1159 - 1 as a comparison star .
its magnitudes of @xmath26 and @xmath27 are quoted from @xcite and @xmath28 , @xmath29 and @xmath30 are from the 2mass catalog @xcite .
we estimated the magnitude of the comparison star at @xmath31 and @xmath32 , using the other nearby stars listed in @xcite .
in addition to the trispec / kanata observation , we performed optical ccd photometry of v2362 cyg from 2006 april 5.8 ( jd 2453831.3 ) , around the fist maximum , to 2007 january 23.4 ( jd 2454123.9 ) with several telescopes in 30-cm class . for differential
photometry , we used the nearby stars , tyc 3181 - 1511 - 1 and tyc 3181 - 1401 - 1 taken in the same frames including the nova . for the correction of interstellar extinction , we used @xmath33 , which is an average of the values reported in @xcite , @xcite and @xcite .
we calculated the interstellar extinction in each band using the reddening curve presented in @xcite and @xcite .
figure [ lc ] and [ kanata ] show the light curves of v2362 cyg .
the latter exhibits the light curves after the rebrightening maximum in detail .
( 85mm,80mm)fig1.eps ( 80mm,60mm)fig2.eps the nova reached its first maximum at v=8.2 on 2006 april 5.8 in our data . in this paper , we denote the fiducial time by @xmath34 , in days , where @xmath35 is defined at the first maximum .
we estimated @xmath36 and @xmath37 from our @xmath20-band light curve , where @xmath38 and @xmath39 are the time taken for the nova to fade by 2 and 3 mag from the maximum , respectively .
our @xmath39 is slightly longer than those in other studies ( @xcite ; @xcite ; @xcite ) .
this is because our maximum magnitude @xmath40 is fainter than others ( @xmath41@xmath42 ) due to the sparseness of our observation around the maximum . even in @xmath39
estimated with our data , v2362 cyg is included in the fast nova , as reported in the other studies . in the early decline phase ( @xmath35@xmath0 ) , the nova showed a monotonous decline without a transition phase , as can be seen in figure [ lc ] .
this figure also shows the nir observations reported in @xcite and @xcite .
we estimated the reddening - corrected color @xmath43 and @xmath44 on @xmath45 and @xmath46 , respectively .
the characteristics of the light curve and color were well consistent with those of typical fast novae .
cllc & remark & @xmath34 days & @xmath20-mag + & discovery & @xmath47 & 10.5 + @xmath48 & first maximum & @xmath49 & 8.2 + @xmath50 & @xmath51 & @xmath52 + @xmath53 & @xmath54 & @xmath55 + @xmath56 & onset of rebrightening & @xmath18 & 12.1 + @xmath57 & rebrightening maximum & @xmath1 & 10.0 + @xmath58 & @xmath25 maximum & @xmath59 & 13.3 + the unusual rebrightening started on @xmath60 , as can be seen in figure [ lc ] .
the slope of the light curve gradually increased until @xmath61 when the object reached the apparent maximum at @xmath62 .
our multi - band photometric observations allow us to study the spectral energy distribution ( sed ) of the optical nir regime during the rebrightening .
figure [ sed ] shows the observed seds . in this figure , we draw 6 seds at epoch ( a , b , c , d , e , and f ) which are shown in figure [ sed ] and also indicated in figure [ kanata ] . at epoch
a , the observed sed can be described with a blackbody emission as shown in figure 3 .
we derived the best - fit temperature at @xmath63 .
the sed , hence , indicates that the optical nir flux was dominated by an optically thick pseudophotosphere at the rebrightening maximum . in ordinary novae
, the blackbody - like photospheric emission dominates the optical flux only around the maximum light . in the case of v2362 cyg
, @xcite reported the optical spectrum during the earliest phase , and suggested that its blue spectrum resembled that of an a5a7 supergiant .
this indicates that the optical emission was dominated by a pseudophotosphere with @xmath64@xmath65 at the outburst maximum @xcite .
the photospheric temperature at the rebrightening maximum was , therefore , comparable to or slightly higher than , that at the first maximum , while the rebrightening maximum was fainter .
assuming blackbody emission , we can readily compare the sizes of the emitting region at the initial and rebrightening maxima using the apparent @xmath20-magnitude and the temperatures . as a result
, the diameter of the photosphere at the rebrightening maximum is estimated to be about half of that at the first maximum .
after the rebrightening maximum , the object started a gradual fading in all optical nir bands .
the @xmath66 color just before and after the rebrightening maximum is shown in figure [ color ] .
there was no significant change across the rebrightening maximum , i.e. , in @xmath67228245 .
this suggests that the shape of the sed remained almost constant irrespective of the flux variation across the rebrightening maximum ( @xmath68 ) .
( 80mm,50mm)fig3.eps ( 80mm,60mm)fig4.eps sudden brightenings in @xmath24- and @xmath25-bands were then detected on @xmath69 ( epoch c ) .
as shown in figure [ sed ] , the pseudophotosphere component rapidly decreased , while another nir component increased around the epoch .
the brightening was observed in @xmath23- and @xmath70-bands a few days later . in figure 3 , we plot @xmath71- , @xmath72- , and @xmath73-band data on @xmath74 ( epoch d ) reported by @xcite and also give blackbody spectra with temperatures of 1500 and 1000 k. the 1500 k blackbody spectrum well explains the observed ir sed on @xmath74 .
hence , this ir emission can be interpreted as the thermal emission from dust grains . @xcite and
@xcite reported the dust temperature of @xmath75 on @xmath74 and @xmath76 on @xmath77 , respectively .
the dust shell would be formed , presumably associated with the rebrightening event .
the rapid decrease of the pseudophotosphere component is likely to be connected with the increasing absorption by the dust shell .
after the @xmath25-band maximum on @xmath78 ( epoch e ) , the object faded in all optical nir bands . at @xmath79 ( epoch f ) ,
as shown in figure [ sed ] , the nir emission can be described with the 1000 k blackbody spectrum .
@xcite reported the dust temperature was below @xmath80 k on @xmath81 .
the dust cooling , thus , seems to have started just after the @xmath25-band maximum , probably due to cessation of the dust formation and expansion of the dust shell .
assuming an optically thin dust cloud and a distance to v2362 cyg of @xmath82 kpc , @xcite estimated the mass of carbonaceous dust grains of @xmath83m@xmath6 .
however , they underestimated the dust mass because they used the nir flux at pre - maximum phase ( epoch d ; 7 days before the @xmath25 maximum ) when the bolometric luminosity of the dust cloud ( @xmath84 ) was still 0.4 times as large as the maximum value on epoch e. the uncertainty of the distance should also be considered for the error estimation of the dust mass .
then , we estimate the dust mass of v2362 cyg with our data under the same assumption as @xcite that carbon dusts were formed after the rebrightening .
although the late time observation ( @xmath85 ) suggests that the ir spectrum was not typical for carbonaceous dust emission @xcite , the dominant species could be changed with epochs ( e.g. , qv vul , @xcite ; v705 cas , @xcite ) according to @xcite , the mass of the dust cloud ( @xmath86 ) is derived as @xmath87 \left ( \frac{t_{\rm dust}}{1000\mbox { k } } \right)^{-6 } \mbox{m}_{\solar},\end{aligned}\ ] ] where @xmath88 is the distance to the nova , n is a number of total dust grains and @xmath89 is a temperature of dust grains .
we assume graphite dust grains of a density of @xmath90 and a planck - mean cross section coefficient @xmath91 ( for smaller grains with size of @xmath92 @xmath93 ; @xcite ) . adopting @xmath94
k estimated from the ir data in @xmath74 ( epoch d ) and @xmath95 w@xmath96 in our data at nir maximum ( @xmath78 , epoch e ) , we obtain @xmath97 we note that the deriving mass does not mean the whole mass formed in the ejecta after the rebrightening event , but the mass evaluated at the nir maximum ( @xmath78 , epoch e ) .
@xmath86 highly depends on the distance @xmath88 , although it has not been determined well ; @xmath98 kpc @xcite , @xmath56@xmath99 kpc @xcite , @xmath100@xmath101 kpc @xcite , @xmath102 kpc @xcite and @xmath103@xmath104 kpc @xcite . for the nearest ( @xmath98 kpc ) and
the most distant ( @xmath105 kpc ) cases , the @xmath86 becomes @xmath106 m@xmath8 and @xmath107 m@xmath8 , respectively . even if we take the uppermost value , @xmath86 is comparable to or smaller than those of ordinary dust - forming novae ( @xmath108@xmath109 m@xmath6 ; e.g. @xcite ) . it is noted that the dust mass should be larger if the dust cloud is optically thick as indicated by @xcite . if this is the case , the derived @xmath86 should be the lower - limit and the actual dust mass would be larger .
on the other hand , our nir data suggests that @xmath84 is only @xmath110@xmath111 times as large as the bolometric luminosity of nova ( @xmath112 ) even at the @xmath25 maximum , if we assume @xmath113 erg s@xmath114 with @xmath82 kpc as @xcite .
@xmath84 is calculated by integrations of blackbody seds fitted with @xmath25-band data assuming dust temperatures with a rage of @xmath115@xmath116 k and @xmath88=7.2 kpc .
these facts suggest that the dust cloud does not fully cover the nova ( i.e. , covers by less than 4@xmath117 str ) , if the dust cloud was optically thick .
a patchy cloud is preferable .
@xcite and @xcite detected p - cygni profiles in the optical spectrum of v2362 cyg during its rising phase to the rebrightening maximum ( @xmath118 ) . in conjunction with the existence of the blackbody emission of @xmath119
k , the rebrightening phase is reminiscent of the maximum phase of ordinary novae ( @xcite ) .
we propose that the rebrightening event was caused by the reformed pseudophotosphere by a delayed mass - outflow .
@xcite recently proposed that the energy source of the rebrightening is magnetic reconnections around the pseudophotosphere formed by the first outflow . such a magnetic activity may have accelerated the gas and caused the second outflow in v2362 cyg . as discussed in section 4.1 ,
the mass of the dust grains formed after the rebrightening is likely to be smaller than those in other dust - forming novae .
if the dust grains are formed in the first ejecta cooled predominantly by expansion ( i.e. , classical dust formation mechanism , e.g. , @xcite ) the epoch of the dust formation is expected to be earlier ( @xmath120 d ) and the dust mass should be larger , as in ordinary dust - forming novae .
the ejecta of the second outflow would contain less materials than that of the initial outflow because the peak luminosity of the rebrightening was much lower than the initial peak .
thus , masses are considered to be small for both the dust and the ejecta in the second outflow .
this suggests that the dust grains would be formed preferentially from the material of the second outflow , if the @xcite s scenario that the dust formation would be triggered by the interaction between the second outflow and the initial flow is the case . here
we briefly discuss the nature of the termination of the rebrightening , i.e. , the beginning of the decline on @xmath121 .
we can consider two scenarios ; ( i ) the rebrightening was terminated due to the increasing absorption for the optical nir flux by the dust grains , and ( ii ) the rebrightening was terminated by a decay of the emission from the pseudophotosphere itself .
if the former scenario is ture , we could expect the reddening of the object just after the rebrightening maximum . as shown in figure [ color
] , however , no reddening was detected for the first 4 d just after the rebrightening maximum .
the observation , hence , favors the latter scenario .
if scenario ( ii ) is the case , the dust grains are likely to begin to be formed about a week after the rebrightening maximum .
this implies that the dust formation might be promoted by the decay of the emission from the shrinking pseudophotosphere . the optical steep decline with the nir brightening after @xmath122 suggests that the increasing absorption due to dust formation is significant .
however , the shrinking pseudophotosphere might also have partly caused the decline of the optical flux ( and a part of nir one ) . a full analysis with a radiation transfer for the pseudophotosphere and the dusty ejecta would be required to reproduce the observed light curves , which is out of the scope of this paper .
we carried out optical and nir photometry of nova v2362 cyg for @xmath123 days from the first maximum .
our observation indicates the re - appearance of pseudophotosphere with @xmath119 k around the rebrightening maximum .
the nir flux began to increase a week after the rebrightening maximum , accompanied with the steep decline of the optical flux .
the @xmath25-band flux peaked at @xmath78 .
the nir flux at the peak is well explained by the thermal dust emission with @xmath124 k. the dust temperature then decreased from @xmath116 k to @xmath125 k within 40 days .
we estimate the dust mass of @xmath126@xmath127 m@xmath8 .
this support that the dust would be formed preferentially from the material of the second outflow , if the dust formation would be triggered by the interaction between the second flow and initial one as suggested in @xcite .
+ + this work was partly supported by a grand - in - aid from the ministry of education , culture , sports , science , and technology of japan ( 19740104 , 17340054 ) . | we present optical and near - infrared ( nir ) photometry of a classical nova , v2362 cyg (= nova cygni 2006 ) .
v2362 cyg experienced a peculiar rebrightening with a long duration from @xmath0 to @xmath1 d after the maximum of the nova .
our multicolor observation indicates an emergence of a pseudophotosphere with an effective temperature of @xmath2 k at the rebrightening maximum . after the rebrightening maximum , the object showed a slow fading homogeneously in all of the used bands for one week .
this implies that the fading just after the rebrightening maximum ( @xmath3 week ) was caused by a slowly shrinking pseudophotosphere .
then , the nir flux drastically increased , while the optical flux steeply declined .
the optical and nir flux was consistent with blackbody radiation with a temperature of @xmath4 during this nir rising phase .
these facts are likely to be explained by dust formation in the nova ejecta . assuming an optically thin case
, we estimate the dust mass of @xmath5 m@xmath6 , which is less than those in typical dust - forming novae .
these results support the senario that a second , long - lasting outflow , which caused the rebrightening , interacted with a fraction of the initial outflow and formed dust grains . |
a standard scenario in quantum parameter estimation can be described as follows : firstly , a probe system would be prepared in an appropriate initial state @xmath116 , and then it undergoes an evolution which would imprint the parameter information onto the evolved state , say @xmath40 , and finally it would subject to a povm measurement . the overall process is repeated @xmath117 times , and we infer the parameter @xmath38 from the statistics of the measurement outcomes by choosing an unbiased estimator .
the variance of this estimator , i.e. , @xmath118 , quantifies the error on estimation of @xmath38 , and is lower bounded by : @xmath119 where @xmath120 is the classical fisher information optimized over all the possible estimators , and qfi is the quantum fisher information , which is further optimized over all the allowable measurements and is given by @xcite @xmath121,\ ] ] where the symmetric logarithmic derivative @xmath122 in the above equation is defined as : @xmath123 writing @xmath40 in its spectral decomposition as @xmath124 , one can obtain @xcite : @xmath125
in this section , we would derive the approximate expressions of qfi for an arbitrary initial state of rp with and without the oscillating field , i.e. , eq .
( 7 ) and eq .
( 9 ) in the main text , respectively .
when the horizontal hf coupling @xmath44 , the role of nuclear spin can be considered as applying an effective magnetic field ( depending on its state ) on the electronic spin . if the nucleus is in the spin up ( down ) state , the effective magnetic field is @xmath126 , with @xmath127 being the @xmath112 direction . as a result , the effective hamiltonian of rp can be written as @xmath128 , where @xmath129 is the geomagnetic field around the rp , with @xmath47 being the intensity of the geomagnetic field , and @xmath48 and @xmath49 being the orientation of the geomagnetic field to the basis of the hf tensor .
the axial symmetry of the hf tensor allows us to set @xmath50 and focus on @xmath48 in the range @xmath51 $ ] without loss of generality , and @xmath48 is the parameter to be estimated for avian compass . here
we denote the eigenstates of the effective hamiltonian @xmath130 as @xmath131 and its corresponding eigenvalues as @xmath132 ( @xmath133 ) .
specifically , @xmath134 and @xmath135 are the eigenstates of hamiltonian of electron 1 , i.e. , @xmath136 . @xmath137 and @xmath138 are the eigenstates of hamiltonian of electron 2 , i.e. , @xmath139 , with @xmath140 , @xmath141 , @xmath142 , @xmath143 , and @xmath144 @xcite . given an arbitrary initial state of rp @xmath54 , we would derive the approximate expression of qfi of the steady state @xmath43 of rp ( see eq .
( 4 ) in the main text ) without considering the oscillating field .
we can always expand @xmath54 in the eigenbasis of the effective hamiltonian @xmath130 as @xmath145 with @xmath146 . generally , the nucleus is initially in a complete mixed state , i.e. , @xmath21 . as a result ,
the state dependent effective magnetic field @xmath126 induced by the nuclear spin leads to the effective hamiltonian of rp @xmath147 with the same probability @xmath12 .
after some calculations , we can obtain the rp density matrix at time @xmath16 analytically : @xmath148 with @xmath149 in most previous studies of avian compass , the recombination rate @xmath30 is generally considered to be the order of @xmath53 . and
in this regime , @xmath150 , thus the high - frequency oscillating terms of eq .
( [ rhot ] ) have no contribution to the time integral of eq . ( 4 ) in the main text , hence the steady state of rp can be expressed as @xmath151 now we consider the strong hf coupling approximation , i.e. , @xmath152 , and expand the eigenvectors @xmath153 and @xmath154 in a power series of @xmath155 , keeping terms to the first order . through our calculation , we obtain that @xmath156 and @xmath157 . submitting them into eq .
( [ rhobar1 ] ) and keeping terms to the first order of @xmath155 , @xmath158 can be approximately simplified as a diagonal form : @xmath159 with @xmath57 , and @xmath58 . and then according to eq .
( [ qqfi ] ) , the qfi of @xmath158 ( eq . ( [ rhobar2 ] ) ) can be obtained analytically : @xmath56^{2}\bigl(\frac{1}{\rho_{i}^{11}}+\frac{1}{\rho_{i}^{22}}\bigr ) + \frac{(\rho_{i}^{11}-\rho_{i}^{22})^{2}}{\rho_{i}^{11}+\rho_{i}^{22}},\ ] ] where @xmath64 $ ] represents the real part of @xmath65 .
now we would derive the approximate expression of qfi of the steady state @xmath43 ( see eq .
( 4 ) in the main text ) for an arbitrary initial state of rp with a weak resonant oscillating field @xmath160 , where @xmath73 is the strength of oscillating field with frequency @xmath74 being resonant with electron 1 .
@xmath75 and @xmath76 represent the direction of oscillating field with respect to the basis of the hf tensor . due to the axial symmetry of the hf tensor we set @xmath77 . here
we consider @xmath78 , namely , the weak oscillating field is perpendicular to earth s magnetic field . for the convenience of our calculation below , we express an arbitrary initial state of rp @xmath54 as @xmath161 with @xmath162 representing the operator of electron 1 .
because of the effect of nucleus , the larmor frequency of electron 2 induced by the effective magnetic field and the geomagnetic field is always not resonant with the frequency of oscillating field , as a consequence , electron 2 can be considered as almost not influenced by the oscillating field @xcite .
based on this , the rp density matrix at time @xmath16 can be obtained as @xmath163 with @xmath164 where @xmath165 are the eigenvalues of @xmath139 , and @xmath166 $ ] is the evolution operator of electron 1 with @xmath167 , and @xmath168 denoting the chronological time - ordering operator . after performing the rotating - wave approximation ,
the evolution operator can be obtained in the eigenbasis @xmath61 ( @xmath10 ) of @xmath63 @xcite : @xmath169 with @xmath170 , and @xmath171 .
when @xmath30 is in the regime @xmath53 , @xmath172 , @xmath173 , thus the high - frequency oscillating terms of eq .
( [ rhott ] ) have no contribution to the time integral of eq . ( 4 ) in the main text , hence the steady state of rp under the influence of oscillating field can be expressed as @xmath174 with @xmath175 where @xmath176 \\ & - \frac{\omega^2}{2(k^2 + \omega^2)}\bigl(\varrho_{\pm}^{11}(1,1 ) - \varrho_{\pm}^{11}(2,2)\bigr ) , \end{split}\ ] ] @xmath177 \\ & - \frac{\omega^2}{2(k^2 + \omega^2)}\bigl(\varrho_{\pm}^{22}(1,1 ) - \varrho_{\pm}^{22}(2,2)\bigr ) , \end{split}\ ] ] @xmath178 \\ & + \frac{\omega^2}{2(k^2 + \omega^2)}\bigl(\varrho_{\pm}^{11}(1,1 ) -\varrho_{\pm}^{11}(2,2)\bigr ) , \end{split}\ ] ] @xmath179 \\ & + \frac{\omega^2}{2(k^2 + \omega^2)}\bigl(\varrho_{\pm}^{22}(1,1 ) - \varrho_{\pm}^{22}(2,2)\bigr ) \end{split}\ ] ] with @xmath180 , ( @xmath181 ) . considering the strong hf coupling approximation , i.e. , @xmath152 , eq .
( [ rr ] ) can be approximately simplified as a diagonal form : @xmath182 where @xmath183 with @xmath82 having been defined below eq .
( [ rhobar2 ] ) , @xmath83 $ ] ( @xmath84 , @xmath85 ) , and @xmath86 $ ] represents the imaginary part of @xmath65 . and
then according to eq .
( [ qqfi ] ) , the qfi of @xmath158 ( eq . ( [ ab ] ) ) can be obtained analytically : @xmath184 ^ 2}{(k^2+\omega^2)^2}\bigl(\frac{1}{p_{i}^{11}}+\frac{1}{p_{i}^{22}}\bigr)+\frac{(p_{i}^{11}-p_{i}^{22})^2}{p_{i}^{11}+p_{i}^{22}}.\ ] ] through our calculation , we obtain that when @xmath87 ( without the oscillating field ) , eq .
( [ qqfii ] ) reduces to eq .
( [ qfia ] ) ; when @xmath88 , qfi@xmath890 , which implies that when the order of @xmath30 is much smaller than that of @xmath90 , the weak resonant oscillating field can completely disorient the bird .
in fact , the order of @xmath30 has been widely accepted to be approximately @xmath93@xmath25@xmath70 as mentioned above . in what follows , without making any approximation ,
we reconsider the order of @xmath30 in terms of qfi , by considering a weak resonant oscillating field of strength brf=150nt perpendicular to earth s magnetic field , which can completely disorient the bird @xcite . here
we also take the singlet state @xmath92 as the initial state of rp as an example .
our numerical results are shown in fig .
a1 , and it can be seen that when @xmath185 , the qfi is almost immune to the oscillating field , and when @xmath69 , the qfi with the oscillating field is reduced to some extent compared with that without the oscillating field , but we are not sure whether this reduction of qfi can disrupt the birds or not . however , when @xmath91 , the qfi with the oscillating field reduces significantly .
thus it is safe to say that if the oscillating field is to disorient the bird , it might be approximately @xmath91 , which is consistent with the previous works @xcite .
below , we would further numerically show that when @xmath91 , for an arbitrary initial state of rp , a weak resonant oscillating field orthogonal to the geomagnetic field can highly reduce the value of qfi of the steady state @xmath43 . specifically , we randomly sample 100 initial states and plot in fig .
a2 the corresponding percent decreases of qfi , i.e. , @xmath186 , as a function of @xmath48 . the results without making any approximation show that for all the sampled initial states
, @xmath187 is larger than @xmath94 , which implies that a weak resonant oscillating field orthogonal to the geomagnetic field can completely disorient the bird for an arbitrary initial state of rp .
the effect of external magnetic field and oscillating field on the value of qfi for @xmath188 has been discussed in the main text . here
we consider the case @xmath189 , and calculate the corresponding qfi . in this case ,
an approximately analytical expression of qfi can not be obtained , thus we calculate the qfi numerically . through our large numerical calculations , we find that our results are not quite sensitive to what the value of hf coupling is . here
we consider @xmath190 t , @xmath191 and the initial state of rp to be the singlet state @xmath92 as an example .
the numerical results are shown in fig .
a3 , and we can see that the qfi of @xmath67 weaker ( @xmath192 t ) and stronger ( @xmath193 t ) fields do almost not change compared with that of geomagnetic field ( @xmath194 t ) , but is highly reduced when a weak resonant oscillating field perpendicular to earth s magnetic field is applied . besides , through our numerical calculations , we find that there is no effect at such weak fields when the oscillating field is parallel to earth s magnetic field .
these results are similar to that without considering the horizontal hf coupling components in the main text .
due to the quantum mechanical nature of rp mechanism , the effect of entanglement on the avian compass has been investigated in terms of the singlet yield @xcite . in this letter
, we reconsider the effect of entanglement on the avian compass in terms of qfi , and use concurrence @xcite to quantify entanglement . the concurrence @xmath195 of two qubits
is defined as @xmath196 , where @xmath197 are the eigenvalues of the matrix @xmath198 arranged in decreasing order , and @xmath43 is defined in eq .
( 4 ) in the main text . as an example
, we plot the qfi and @xmath199 as functions of the recombination rate @xmath30 for @xmath200 t , @xmath201 , @xmath190 t , and @xmath188 with the rp initial state being the singlet state @xmath92 in fig . a4 .
and we can see from fig .
a4 that entanglement can not help to promote bird orientation , to be more specific , when @xmath30 is smaller , the qfi is relatively larger which actually corresponds to zero entanglement , and when @xmath30 is larger , the qfi is reduced to zero which corresponds to a relatively larger entanglement instead .
it is noted that similar conclusions can be obtained for any other direction angles through our large numerical calculations .
in fact , the behavior of entanglement as a function of @xmath30 can be also seen from the expression of @xmath43 .
specifically , when @xmath30 is small , @xmath43 becomes a separable state ( see eq .
( [ rhobar1 ] ) ) , which implies that there is no entanglement in @xmath43 as shown in fig . a4 .
however , when @xmath30 is large , it can be derived from eq .
( [ rhot ] ) and eq .
( [ rho3 ] ) that @xmath43 becomes the singlet state , because in this case the lifetime of rp ( @xmath202 ) is too short to make a transition between the singlet and triplet states , so the concurrence @xmath199 equals to 1 as shown in fig . a4 .
decoherence is unavoidable for the rp , and now we reconsider its effect on the avian compass in terms of qfi , with the singlet state @xmath92 being the initial state of rp as an example .
specifically , we display three typical classes of independent markovian environmental noise , namely , the amplitude damping noise , dephasing noise and depolarized noise .
we describe the environmental noises by the standard lindblad master equation : @xmath203+\sum_{i}\gamma_{i}\bigl(l_{i}\rho(t)l_{i}^{\dag}-\frac{1}{2}\{l_{i}^{\dag}l_{i},\rho(t)\}\bigr),\ ] ] where @xmath204 ( see eq . ( 1 ) in the main text ) denotes the total hamiltonian for each rp , @xmath205 represents the density matrix of one nucleus and two electrons at time @xmath16 , @xmath206 represents the decoherence rate , @xmath207 represents the anticommutator , and @xmath208 is the lindblad operator . for the amplitude damping noise , @xmath208 is only @xmath209 for each electronic spin individually ( i.e. , tensored with identity matrices for the nuclear spin and the other electronic spin ) ; for the dephasing noise , @xmath208 is only @xmath62 for each electronic spin individually ; for the depolarized noise , @xmath208 are @xmath210 , @xmath211 , @xmath62 for each electronic spin individually .
firstly , let us examine the effect of uncorrelated amplitude damping noise , with the numerical results shown in fig .
and we can see from fig .
a5(a ) that when @xmath212 , the qfi of @xmath67 weaker and stronger fields are almost not changed compared with that of geomagnetic field . moreover , a weak resonant oscillating field @xmath213nt , which is perpendicular to earth s magnetic field , can highly reduce the qfi with @xmath200 t , and the percent decrease of qfi , i.e. , @xmath187 , can be larger than @xmath214 shown in the inset of fig .
a5(a ) , which is large enough to imply that a weak oscillating field can completely disrupt the birds . and for @xmath215 in fig .
a5(b ) , the @xmath67 stronger and weaker fields still have little influences on the value of qfi .
meanwhile , there still exists an obvious difference in the value of qfi with and without the oscillating field , with @xmath187 being larger than @xmath216 shown in the inset of fig .
however , when @xmath217 in fig .
a5(c ) , we can see that although the curves of qfi for different magnetic field intensities overlap completely , it would render the bird almost immune to the weak oscillating field , with @xmath187 being smaller than @xmath218 shown in the inset of fig .
a5(c ) , which can not account for the fact that a weak oscillating field can completely disrupt the avian compass . as a conclusion , the decoherence rate should be smaller than @xmath219 for this amplitude damping noise .
next , we consider the effect of uncorrelated dephasing noise , with the numerical results shown in fig . a6 . from fig .
a6 we can see that the @xmath67 stronger and weaker fields compared with the geomagnetic field have almost no influences on the value of qfi for @xmath220 and @xmath221 . however , the effects of a resonant oscillating field on the value of qfi are different . specifically , from fig .
a6(a ) , we can see that when @xmath212 , the qfi would be highly reduced when the oscillating field is applied , with @xmath187 being larger than @xmath214 shown in the inset of fig .
a6(a ) . and
when @xmath215 , the oscillating field is still able to reduce the value of qfi to some extent , especially when @xmath48 is small , @xmath187 can reach approximately @xmath222 shown in the inset of fig .
but when @xmath217 , we can see from fig .
a6(c ) that bird becomes quite immune to the weak oscillating field with @xmath187 being approximately equal to 0 for large @xmath48 .
thus for this dephasing noise , the decoherence rate should be smaller than @xmath219 .
finally , we consider the effect of uncorrelated depolarized noise , with the numerical results shown in fig .
a7 we can see that when @xmath212 , the qfi of @xmath67 weaker and stronger fields are almost unchanged compared with that of geomagnetic field , and the difference in the value of qfi with and without the oscillating field is obvious , with the percent decrease @xmath187 being larger than @xmath223 shown in the inset of fig .
a7(a ) . however , when @xmath224 , on the one hand , the value of qfi is significantly small despite of its insensitivity to the @xmath67 weaker and stronger fields . on the other hand ,
the oscillating field has almost no effect on the value of qfi , with @xmath187 being approximately @xmath225 shown in the inset of fig .
a7(b ) or smaller than @xmath226 shown in the inset of fig .
a7(c ) . as a result ,
for this uncorrelated depolarized noise , the decoherence rate should be smaller than @xmath227 .
s. lloyd , a quantum of natural selection , nature phys . * 5 * , 164 ( 2009 ) .
j. c. brookes , f. hartoutsiou , a. p. horsfield , and a. m. stoneham , could humans recognize odor by phonon assisted tunneling ?
98 * , 038101 ( 2007 ) .
i. a. solovyov , p. y. chang , and k. schulten , vibrationally assisted electron transfer mechanism of olfaction : myth or reality ?
* 14 * , 13861 ( 2012 ) .
z. d. nagel , j. p. klinman , tunneling and dynamics in enzymatic hydride transfer .
chem . rev . * 106 * , 3095 ( 2006 ) .
r. k. allemann , n. s. scrutton , quantum tunnelling in enzyme - catalysed reactions ( rsc , 2009 ) . m. b. plenio , and s. f. huelga , dephasing - assisted transport : quantum networks and biomolecules , new j. phys . * 10 * , 113019 ( 2008 ) .
m. mohseni , p. rebentrost , s. lloyd and a. a. guzik , environment - assisted quantum walks in photosynthetic energy transfer , j. chem .
phys . * 129 * , 174106 ( 2008 ) .
k. maeda , k. b. henbest , f. cintolesi , i. kuprov , c. t. rodgers , p. a. liddell , d. gust , c. r. timmel , and p. j. hore , chemical compass model of avian magnetoreception , nature ( london ) * 453 * , 387 ( 2008 ) . c. t. rodgers and p. j. hore , chemical magnetoreception in birds : the radical pair mechanism , proc .
sci . * 106 * , 353 ( 2009 ) .
k. schulten , c. e. swenberg , and a. weller , a biomagnetic sensory mechanism based on magnetic field modulated coherent electron spin motion , z. phys . chem .
* nf111 * , 1 ( 1978 ) .
t. ritz , s. adem , and k. schulten , a model for photoreceptor - based magnetoreception in birds , biophys .
j. * 78 * , 707 ( 2000 ) .
j. a. pauls .
y. zhang , g. p. berman , s. kais , quantum coherence and entanglement in the avian compass , phys .
e * 87 * , 062704 ( 2013 ) . j. cai , g. g. guerreschi , and h. j. briegel , quantum control and entanglement in a chemical compass , phys .
rev . lett . * 104 * , 220502 ( 2010 ) .
j. cai and m. b. plenio , chemical compass model for avian magnetoreception as a quantum coherent device , phys .
lett . * 111 * , 230503 ( 2013 ) . h. j. hogben , t. biskup , and p. j. hore ,
entanglement and sources of magnetic anisotropy in radical pair - based avian magnetoreceptors , phys .
* 109 * , 220501 ( 2012 ) .
j. n. bandyopadhyay , t. paterek , and d. kaszlikowski , quantum coherence and sensitivity of avian magnetoreception , phys .
109 * , 110502 ( 2012 ) .
j. cai , f. caruso , and m. b. plenio , quantum limits for the magnetic sensitivity of a chemical compass , phys .
a * 85 * , 040304(r ) ( 2012 ) .
b. m. xu and j. zou , dark state population determines magnetic sensitivity in radical pair magnetoreception model , sci .
* 6 * , 22417 ( 2016 ) .
e. m. gauger , e. rieper , j. j. l. morton , s. c. benjamin , and v. vedral , sustained quantum coherence and entanglement in the avian compass , phys .
* 106 * , 040503 ( 2011 )
. b. m. xu , j. zou , h. li , j. g. li , and b. shao , effect of radio frequency fields on the radical pair magnetoreception model , phys .
e * 90 * , 042711 ( 2014 ) .
i. k. kominis , quantum zeno effect explains magnetic - sensitive radical - ion - pair reactions , phys .
e * 80 * , 056115 ( 2009 ) .
l. p. yang , q. ai , and c. p. sun , generalized holstein model for spin - dependent electron - transfer reactions , phys .
a * 85 * , 032707 ( 2012 ) c. y. cai , q. ai , h. t. quan , and c. p. sun , sensitive chemical compass assisted by quantum criticality , phys .
a * 85 * , 022315 ( 2012 ) .
a. imamoglu , k. b. whaley , photoactivated biological processes as quantum measurements , phys .
e * 91 * , 022714 ( 2015 ) .
m. tiersch and h. j. briegel , decoherence in the chemical compass : the role of decoherence for avian magnetoreception , phil .
a * 370 * , 4517 ( 2012 ) .
a. chia , k. c. tan , .
pawela , p. kurzyski , t. paterek , and d. kaszlikowski , coherent chemical kinetics as quantum walks .
i. reaction operators for radical pairs , phys .
e * 93 * , 032407 ( 2016 ) .
k. mouloudakis , i. k. kominis , quantum information processing in the radical - pair mechanism , arxiv:1607.03071 ( 2016 ) .
n. lambert , y. n. chen , y. c. cheng , c. m. li , g. y. chen and f. nori , quantum biology , nature phys .
* 9 * , 10 ( 2012 ) .
m. arndt , t. juffmann , v. vedral , quantum physics meets biology , hfsp j. * 3 * , 386 ( 2009 ) .
s. f. huelgaa and m. b. plenio , vibrations , quanta and biology , contemp . phys .
* 54 * , 181 ( 2013 ) .
w. wiltschko and r. wiltschko , magnetic compass of european robins , science * 176 * , 62 ( 1972 ) .
w. wiltschko , _ in animal migration , navigation , and homing , edited by k. schmidt - koenig and w. t. keeton _
( springer , new york , 1978 ) , p. 302 .
w. wiltschko , k. stapput , p. thalau , and r. wiltschko , avian magnetic compass : fast adjustment to intensities outside the normal functional window , naturwissenschaften * 93 * , 300 ( 2006 ) .
m. winklhofer , e. dylda , p. thalau , w. wiltschko , and r. wiltschko , avian magnetic compass can be tuned to anomalously low magnetic intensities , proc .
b * 280 * , 20130853 ( 2013 ) .
t. ritz , p. thalau , j. b. phillips , r. wiltschko , w. wiltschko , resonance effects indicate a radical - pair mechanism for avian magnetic compass , nature ( london ) * 429 * , 177 ( 2004 ) .
p. thalau , t. ritz , k. stapput , r. wiltschko , w. wiltschko , magnetic compass orientation of migratory birds in the presence of a 1.315 mhz oscillating field , naturwissenschaften * 92 * , 86 ( 2005 ) .
t. ritz , r. wiltschko , p. j. hore , c. t. rodgers , k. stapput , p. thalau , c. r. timmel , w. wiltschko , magnetic compass of birds is based on a molecule with optimal directional sensitivity , biophys .
j. * 96 * , 3451 ( 2009 ) . v. giovannetti , s. lloyd , and l. maccone , quantum metrology , phys .
* 96 * , 010401 ( 2006 ) .
v. giovannetti , s. lloyd , and l. maccone , advances in quantum metrology , nature photon . * 5 * , 222 ( 2011 ) .
m. a. taylor , w. p. bowen , quantum metrology and its application in biology , phys . rep . * 615 * , 1 ( 2016 ) .
u. e. steiner and t. ulrich , magnetic field effects in chemical kinetics and related phenomena , chem . rev . * 89 * , 51 ( 1989 ) .
h. cramr , _ mathematical methods of statistics _ , vol .
9 ( princeton university press , 1999 ) . s. l. braunstein and c. m. caves , statistical distance and the geometry of quantum states , phys .
lett . * 72 * , 3439 ( 1994 ) . c. w. helstrom , _ quantum detection and estimation theory _ , ( academic press , new york , 1976 ) .
c. w. helstrom , _ probabilistic and statistical aspects of quantum theory _ , ( north holland , amsterdam , 1982 )
. m. o. scully and m. s. zubairy , _ quantum optics _ ,
2 ( cambridge university press , 1997 ) , p. 151 . y. zhang , g. p. berman , s. kais , sensitivity and entanglement in the avian chemical compass , phys . rev .
e * 90 * , 042707 ( 2014 ) .
w. k. wootters , entanglement of formation of an arbitrary state of two qubits , phys .
lett . * 80 * , 2245 ( 1998 ) . | the radical pair ( rp ) mechanism , which describes the quantum dynamics of a spatially separated electron pair , is considered as one of the principal models of avian magnetoreception .
different from the conventional phenomenological approach where the sensitivity of avian magnetoreception is characterized by the singlet yield @xmath0 , we introduce the quantum fisher information ( qfi ) , which represents the maximum information about the magnetic field s direction extracted from the rp state , to give a precise measure of sensitivity of avian compass essentially .
the consistency between our results and experimental observations suggests that qfi plays a decisive role in avian magnetoreception .
besides , within the framework of quantum metrology , we can judge the feasibility of any possible measurement scheme for avian magnetoreception , and shed light on an intrinsic relevance between the singlet yield and a concrete measurement scheme of our approach .
the present work allows us to understand many things about avian magnetoreception from a fully new perspective of quantum metrology , and provide a new route to establish a direct connection between quantum information and many other biological functions . _
introduction_. recent evidence suggests that some unique features of quantum mechanics can be harnessed to enhance biological functions in a large variety of living organisms , e.g. , in natural selection @xcite , olfaction sense @xcite , enzymatic reactions @xcite , photosynthetic light harvesting @xcite , avian magnetoreception @xcite , etc .
, which indicates that quantum biology has been entering a new stage @xcite .
as one of the principal models of avian magnetoreception , the radical pair ( rp ) mechanism @xcite , based on singlet - triplet transitions due to the anisotropic hyperfine ( hf ) interaction , suggests that migratory birds depend on the photoinduced rps for navigation , which has been supported by intensive evidences and behavioral experiments with birds @xcite . due to the quantum mechanical nature of rp model , a growing interest in understanding the function of avian magnetoreception has extended from chemists , biologists to physicists , by using the rich fruits in the field of quantum information such as quantum coherence and entanglement @xcite .
however , a deeper understanding of the mechanism of avian magnetoreception may need the ability to precisely measure the function of avian compass . with the development of various kinds of quantum techniques , particularly in the field of quantum metrology @xcite , which has primarily been developed to find the fundamental limit to precision of estimating an unknown parameter ,
can we use the method of quantum metrology to precisely characterize the sensitivity of avian magnetoreception ? in this letter , we apply quantum metrology to avian magnetoreception , and use the quantum fisher information ( qfi ) to give a precise measure of sensitivity of avian magnetoreception , which represents the maximum information about the geomagnetic field direction extracted from the rp state . such an approach allows us to establish a quantitative connection between the performance of avian compass and the magnitude of qfi .
although there have been a few works which noticed the potential relevance between quantum metrology and biology @xcite , we have not see any relevant works to date which really characterized the magnetic sensitivity of avian compass by means of quantum metrology . in the context of rp model ,
we first derive a statistical average state ( i.e. , a steady state ) of rp , then calculate the qfi of this state and finally compare the results with the relevant experimental results . the highly consistency between the behavior of qfi and the experimental results
underlies a decisive role played by the qfi in avian magnetoreception .
besides , within the framework of quantum metrology , we can judge the feasibility of any possible measurement scheme for avian magnetoreception , and shed light on an intriguing connection between the conventional approach ( i.e. , the singlet yield ) @xcite and a concrete measurement scheme of our approach .
_ rp model and its statistical average state_. in the avian compass , each photoinduced rp has a spatially separated electron pair coupled to an external magnetic field @xmath1 and a few nuclei .
generally it is believed that only one of the electrons interacts with the nuclei with an anisotropic hf coupling and the other is free @xcite .
thus this provides asymmetry and leads to singlet - triplet transition required for the directional sensitivity . in this letter
, we only consider the simple case of one nuclear spin and the corresponding hamiltonian for each rp is @xcite @xmath2 where @xmath3 is the external magnetic field around the rp , @xmath4 is the gyromagnetic ratio , with @xmath5 being the bohr s magneton and @xmath6 being the @xmath7 factor of electron . here
, we assume that the @xmath7 factor is the same for both electronic spins and set its value according to free electron , i.e. , @xmath8 .
@xmath9 are the electronic spin operators ( @xmath10 ) , and @xmath11 is the nuclear spin @xmath12 operator .
@xmath13 is the hf tensor which couples the nuclear spin and electron 2 with a diagonal form @xmath14 , and we assume an axially symmetric ( or cigar - shaped ) hf tensor , i.e. , @xmath15 .
the rp density matrix at time @xmath16 can then be described as @xmath17,\ ] ] where @xmath18 is the evolution operator corresponding to the hamiltonian eq .
( [ h ] ) , and @xmath19 $ ] means taking the trace over the nucleus .
@xmath20 is the initial state of two electrons and one nucleus , and generally the nucleus is initially in a complete mixed state , i.e. , @xmath21 .
first we assume that the rps are identical and in the same initial state .
due to the continuous optical excitation , the creation of each rp is entirely accidental and its decay is also random .
however , with respect to all the rps existing in the bird s eye , they would be in a steady state . in what follows
, we would derive a statistical average state of rp to describe this steady state . to be more specific ,
choosing an arbitrary fixed time to see ( here we set the fixed time as the reference time , denoted as @xmath22 ) , the rps at the reference time ( @xmath22 ) are constituted of those evolved from different time @xmath23 ( @xmath24 ) , i.e. , the moment of rp formation .
it is reasonable to assume that in time regime @xmath23@xmath25@xmath26 , the number of rps created by optical excitation is a constant which is not dependent on the specific time @xmath23 , denoted by @xmath27 . and
the number of them which still exist ( not decay ) at the reference time is @xmath28 , where @xmath29 , with @xmath30 being the recombination rate @xcite . in other words , for each rp created by optical excitation in time regime @xmath23@xmath25@xmath26 , its existing probability at the reference time is @xmath31 and the corresponding state at the reference time ( @xmath22 ) is described as @xmath32 which is evolved from the time regime @xmath23@xmath25@xmath26 . due to the fact that each rp is subject to the optical excitation randomly , at the reference time
, the state of the rp would be consisted of a large number of states evolved from different time @xmath23 with a corresponding weight @xmath33 . as a result ,
we can obtain a statistical average state ( i.e. , the steady state ) of rp : @xmath34 where in the second equation , we have replaced the integration variable @xmath23 with @xmath35 , and accordingly @xmath32 is equal to @xmath36 defined in eq .
( [ rho ] ) .
here it is noted that @xmath37 .
_ magnetic sensitivity quantified by qfi_. for the avian compass , the estimated parameter is the geomagnetic field orientation to the basis of hf tensor . and
according to the quantum parameter estimation theory ( refer to appendix a for a brief introduction ) , the qfi for estimating an unknown parameter @xmath38 can be obtained as @xcite @xmath39 where @xmath40 is the parameter dependent state , with @xmath41 being its eigenstate and @xmath42 its corresponding eigenvalue . in what follows
we would calculate the qfi of the steady state @xmath43 of rp . in the main text
, we only consider a simple case where @xmath44 ( and the case @xmath45 is considered in appendix c ) .
generally , the geomagnetic field can be described as @xmath46 where @xmath47 is the intensity of the geomagnetic field , and @xmath48 and @xmath49 describe the orientation of the geomagnetic field to the basis of hf tensor .
the axial symmetry of hf tensor allows us to set @xmath50 and focus on @xmath48 in the range @xmath51 $ ] without loss of generality , and @xmath48 is the parameter to be estimated for the avian compass .
and then we can calculate the qfi of the steady state @xmath43 of rp under the influence of the geomagnetic field , and in this case , @xmath52 in eq .
( [ h ] ) . in most previous studies of avian compass ,
the recombination rate @xmath30 is generally considered to be the order of @xmath53 . and
in this regime , for an arbitrary initial state of rp @xmath54 , an approximate expression of qfi of the steady state @xmath43 can be obtained , by making a strong hf coupling approximation , i.e. , @xmath55 ( the detailed derivation of qfi can be seen in appendix b ) : @xmath56^{2}(\frac{1}{\rho_{i}^{11}}+\frac{1}{\rho_{i}^{22}})+\frac{(\rho_{i}^{11}-\rho_{i}^{22})^{2}}{\rho_{i}^{11}+\rho_{i}^{22}},\ ] ] where @xmath57 , and @xmath58 , with @xmath59 ( @xmath60 ) and @xmath61 ( @xmath10 ) being the eigenstates of @xmath62 of electron 2 and hamiltonian of electron 1 , i.e. , @xmath63 , respectively , and @xmath64 $ ] represents the real part of @xmath65 . from eq .
( [ qfia ] ) we can see that for any given initial state @xmath54 , the qfi of the steady state @xmath43 is not dependent on @xmath47 , which implies that the change of the intensity of external magnetic field @xmath47 would not disorient the bird permanently .
this is consistent with the experimental result that bird can adapt to different magnetic field intensities @xcite .
furthermore , without making any approximation , we numerically plot the qfi for different magnetic field intensities with the rp initial state being the singlet state @xmath66 in fig .
1 as an example .
we can see from fig .
1 that the @xmath67 weaker ( 32.2@xmath68 t ) and stronger ( 59.8@xmath68 t ) fields @xcite compared with the geomagnetic field ( 46@xmath68 t ) have almost no influences on the value of qfi , that is , bird would not disorient when the intensity of magnetic field is decreased or increased by about @xmath67 of that of geomagnetic field .
it is noted that for @xmath69 and @xmath70 , the above results also hold .
next , we would investigate the influence of an additional weak resonant oscillating field , and in this case , @xmath71 in eq .
( [ h ] ) with @xmath72 where @xmath73 is the strength of oscillating field with frequency @xmath74 being resonant with the free electron .
@xmath75 and @xmath76 represent the direction of oscillating field with respect to the basis of hf tensor . due to the axial symmetry of hf
tensor we set @xmath77 .
firstly , we consider @xmath78 , i.e. , the weak oscillating field is perpendicular to earth s magnetic field . in this case , when @xmath30 is in the regime @xmath53 , for an arbitrary initial state of rp @xmath54 , we can also obtain an approximate expression of qfi of the steady state @xmath43 , by making a strong hf coupling approximation ( see appendix b for a detailed derivation ) : @xmath79 ^ 2}{(k^2+\omega^2)^2}\bigl(\frac{1}{p_{i}^{11}}+\frac{1}{p_{i}^{22}}\bigr)+\frac{(p_{i}^{11}-p_{i}^{22})^2}{p_{i}^{11}+p_{i}^{22 } } , \end{split}\ ] ] where @xmath80 , @xmath81 , with @xmath82 having been defined below eq .
( [ qfia ] ) , @xmath83 $ ] ( @xmath84 , @xmath85 ) , and @xmath86 $ ] represents the imaginary part of @xmath65 . through our calculation , we obtain that when @xmath87 ( without the oscillating field ) , eq .
( [ qfibrf ] ) reduces to eq .
( [ qfia ] ) ; when @xmath88 , qfi@xmath890 , which implies that when the order of @xmath30 is much smaller than that of @xmath90 , the weak resonant oscillating field can completely disorient the bird .
it is noted that the present conclusions are also not dependent on the specific kind of initial state of rp .
furthermore , without making any approximation , for @xmath91 , we numerically plot the qfi with the weak resonant oscillating field orthogonal to the geomagnetic field in fig .
1 , for the rp initial state being the singlet state @xmath92 as an example .
we can see from fig .
1 that the value of qfi is highly reduced when the weak resonant oscillating field is applied , which satisfies the experimental result that a weak resonant oscillating field can disrupt the bird completely @xcite . besides , we find that the oscillating field parallel to earth s magnetic field does not affect the value of qfi which is consistent with the experimental results @xcite . here
we emphasize that in appendix b , we not only discuss the order of @xmath30 in terms of qfi , and obtain that @xmath30 should be the order of @xmath93 , which is in accordance with the previous works @xcite , but also numerically show that when @xmath91 , for an arbitrary initial state of rp , a weak resonant oscillating field orthogonal to the geomagnetic field can reduce the value of qfi by at least @xmath94 of that without the oscillating field . _
discussionpossible implementations for avian compass_. in this letter , we use the qfi to quantify the magnetic sensitivity of avian compass , but it is only an upper bound of precision for magnetoreception .
in fact , there may exists several possible implementations , and what specific kind of implementation is adopted by birds in nature is not clear for us , despite of the prevailing view that the external magnetic field information can be recorded by the singlet yield @xcite which can be detected by birds .
given that the initial state of rp is in the singlet state @xmath92 , we give two possible implementations as examples here , which are the measurement of total angular momentum and that of the square of total magnetic moment .
when a specific povm measurement , corresponding to the observable @xmath95 , has been performed , the unknown parameter @xmath48 can be estimated from the mean value of @xmath95 , with the precision given by the standard error propagation formula @xmath96 @xcite , where @xmath97 and @xmath98 represent the variance and mean value of the observable @xmath95 obtained for @xmath43 , respectively .
firstly , we give the measurement of total angular momentum , i.e. , @xmath99 . here it should be noted that @xmath100 , with @xmath101 representing the probability that the rp is found in the singlet state @xmath92 , besides , it can be seen from eq .
( [ rhoba ] ) that @xmath102 @xcite .
thus , the widely used signal contrast @xmath103 @xcite , which denotes the difference between the maximum and the minimum singlet yields along all the directions , is actually corresponding to the measurement of @xmath104 . moreover ,
through our calculations , we find that @xmath105 is equal to the inverse of the classical fisher information @xmath106 .
although the signal contrast @xmath107 and the classical fisher information f(1/@xmath105 ) both describe the magnetic sensitivity of avian compass , f(1/@xmath105 ) , which denotes the maximum information about @xmath48 extracted from the steady state of rp for this measurement scheme , is more accurate and can better reflect the essence of avian magnetoreception than @xmath107 .
our numerical results of @xmath108 are shown in fig .
2(a ) , and we can see that the @xmath67 stronger and weaker fields compared with earth s magnetic field almost have no influences on the value of @xmath108 , however , a weak resonant oscillating field perpendicular to earth s magnetic field reduces the value of @xmath108 dramatically .
next , we give the measurement of the square of magnetic moment , i.e. , @xmath109 , where @xmath110 and @xmath111 represent the @xmath112 component of spin angular momentum of electron 1 and 2 of rp , respectively .
when the initial state of rp is in the singlet state @xmath92 , @xmath113 . in this case , the measurement of @xmath114 can be considered as a sense of fluctuation of magnetic moment , for @xmath115 .
it is noted that @xmath105 is also equal to the inverse of the classical fisher information @xmath106 through our calculations .
our numerical results of @xmath108 are shown in fig .
2(b ) , and we can see that @xmath108 is robust to different magnetic field intensities , but would be highly reduced when a weak resonant oscillating field is applied .
in fact , in the spirit of this line , we can judge the feasibility of any possible measurement scheme for avian magnetoreception by comparing the corresponding measurement results with the relevant experiment results .
_ other results_. following the present insight that the qfi can well quantify the magnetic sensitivity of avian compass , it is possible to study the effects of entanglement and different decoherence models on the value of qfi in a unified picture .
here we also take the singlet state @xmath92 as the initial state of rp as an example .
firstly , we find that for an arbitrary direction angle @xmath48 , when @xmath30 is small , the qfi is relatively large while the entanglement is equal to 0 , however , when @xmath30 is large , the entanglement becomes large while the qfi reduces to 0 , which implies that entanglement can not help to promote bird orientation ( a detailed discussion can be seen in appendix d ) .
next , we investigate the effects of three typical classes of independent markovian environmental noise on the value of qfi , namely , the amplitude damping noise , dephasing noise and depolarized noise . by comparing our numerical results with the experimental observations , we find that for the amplitude damping noise and the dephasing noise , the decoherence rate should be smaller than 10@xmath30 , while for the depolarized noise , the decoherence rate should even be smaller than @xmath30 ( a detailed discussion can be seen in appendix e )
. _ summary and outlook_. as a precise measure of avian compass , the qfi essentially determines the ability of migratory birds to sense the direction of earth s magnetic field .
compared with the conventional phenomenological approach where the magnetic sensitivity is quantified by the signal contrast @xmath103 , our approach proves more accurate and can better reflect the essence of avian magnetoreception .
meanwhile , in this unified approach of qfi , the order of the recombination rate and the effects of entanglement and decoherence on avian magnetoreception can be well understood .
considering that the qfi is only an upper bound of precision for directional detection , it is desirable to seek for a potential measurement scheme to characterize the compass sensitivity . in the spirit of our approach , we can judge the feasibility of any possible measurement scheme for avian magnetoreception , and have found that the conventional phenomenological approach , i.e , the singlet yield , is just one of the several feasible measurement schemes . then an open question naturally arises : is the widely used singlet yield the optimal choice for describing the sensitivity of avian compass in a practical way ? we hope that our approach can provide a new route to apply the qfi into many other biological processes , such that a precise measure of biological function can be given , and a more profound understanding of biological phenomena can be obtained , which may in turn give us a few clues in the quest to develop quantum technology . _
acknowledgements_. this work was supported by the national natural science foundation of china ( grants no .
11274043 , 11375025 ) .
l. s. guo and b. m. xu contributed equally to this work . |
super - star clusters are extreme among clusters of stars .
they are compact and very luminous , and many are young versions of the massive globular clusters found in giant galaxies like the milky way . the milky way , however , has not been able to form a cluster as compact and massive as a globular cluster for about 10 gy ( although there is a controversial claim that one is forming now
kndlseder 2000 ) . in spite of this ,
six super - star clusters are known in five nearby dwarf irregular ( dim ) galaxies and are inferred to be present , though still embedded , in 4 others . this led billett , hunter , & elmegreen ( 2002 ) to question what conditions allowed these tiny i m galaxies to form such massive clusters .
billett et al .
( 2002 ) undertook a survey of a sample of i m galaxies that had been observed by the _ hubble space telescope_. they searched 22 galaxies for super - star clusters and the less extreme populous clusters .
they found that super - star clusters are actually relatively rare in i m galaxies , but when they form , they seem to be anomalously luminous compared to other clusters in the galaxy .
that is , the super - star clusters in these galaxies are not part of the normal cluster population as they are in spirals ( larsen & richtler 2000 ) .
furthermore , most of the i m galaxies that contain them are interacting with another galaxy or undergoing a starburst , suggesting that special events in the life of the galaxy are required to produce the conditions necessary to form the most massive star clusters .
we were intrigued by the question of where the magellanic clouds would fall in this scheme of cluster formation .
we knew that the lmc contained at least one super - star cluster and that both the lmc and smc contained numerous populous clusters .
therefore , it was not obvious to us that the massive star clusters in these galaxies would stand apart from the rest of the cluster population .
the work of larsen & richtler ( 2000 ) , in fact , suggested that the magellanic clouds follow the correlations set by giant spirals , implying that the formation of massive star clusters is just part of the normal cluster formation process in these galaxies .
surveys of clusters in most i m galaxies are incomplete for all but the most massive star clusters .
the exceptions are the lmc and smc which are close enough for a detailed survey of even faint clusters .
hodge ( 1988 ) predicted that there are of order 4200 clusters in the lmc , and current catalogues list 6659 clusters and associations ( bica et al .
1999 ) . in the smc ,
hodge ( 1986 ) predicted 2000 clusters and the bica & dutra ( 2000 ) catalog contains 1237 clusters and associations .
by contrast , the survey of clusters in ngc 4449 by gelatt , hunter , & gallagher ( 2001 ) yielded 61 objects , yet ngc 4449 is comparable in luminosity to the lmc and so one might expect ngc 4449 to contain thousands of clusters .
most of the clusters in the ngc 4449 survey have m@xmath7 , and the survey was certainly not complete to this magnitude .
in addition , the lmc and smc both contain clusters at the massive end of the spectrum .
therefore , the lmc and smc are the best i m galaxies in which to examine the statistics of the cluster populations . therefore , we set out to answer the question : are the super - star clusters and populous clusters in the magellanic clouds merely the top end of the continuum of clusters , or do they stand apart as anomalous relative to the rest of the cluster population ? to answer this question , we need the mass function of star clusters . however , masses are not known for most of the clusters in the clouds and there is no feasible way of measuring them all directly .
instead , we used the luminosity of the cluster as an indicator of the mass . under the reasonable assumption that all star clusters have formed stars from the same stellar initial mass function , the luminosity is proportional to the mass , and we can substitute the luminosity function for the mass function .
the complication is that clusters fade with time .
therefore , we must compare the luminosities at a fiducial age .
after billett et al .
( 2002 ) , we adopt 10 my as the age at which to compare cluster luminosities .
this , however , means that we must determine the age of each cluster in order to correct the observed luminosity to that at 10 my .
determining the age of each cluster is non - trivial , but doable , and that is what we have done here . in what follows
we discuss the steps that led to the m@xmath8 function and the resulting mass function of star clusters in the lmc and smc .
we used existing catalogues of clusters ; measured ubvr photometry for each cluster ; compared the colors to cluster evolutionary models to determine an age ; corrected the observed m@xmath8 to m@xmath8(10 my ) , the m@xmath8 the cluster would have had at an age of 10 my ; converted m@xmath8(10 my ) to mass , and examined the distribution functions of these quantities for the ensemble and functions of time .
the term `` populous cluster '' was first used by hodge ( 1961 ) to refer to the rich compact clusters in the magellanic clouds .
the use of the term `` super - star cluster '' arose later to emphasize their extreme nature ( van den bergh 1971 ) .
however , these terms had no quantitative definition . for their survey of clusters in i m galaxies , billett et al .
( 2002 ) adopted definitions based on the integrated m@xmath8 of the cluster at the fiducial age of 10 my .
they defined a super - star cluster as a cluster with a magnitude at 10 my of @xmath910.5 or brighter , and , after larsen & richtler ( 2000 ) , they used @xmath10 as the faint limit for populous clusters .
we will adopt these definitions here .
extensive catalogues of star clusters in the magellanic clouds exist in the literature .
most recently , bica & dutra ( 2000 ) have cataloged clusters in the smc , and bica et al .
( 1999 ) , in the lmc .
the optical gravitational lensing experiment ( ogle ) has also produced a catalog of clusters found using visual inspection and automatic algorithms from their imaging materials taken for other purposes ( pietrzyski et al .
1998 , 1999 ) .
these catalogues include previous lists of clusters as well as new candidates found by the authors . these are probably the most complete published catalogues , and so these are what we used . for the smc , we began with the bica and dutra catalog since it already included additional clusters found by the ogle group .
for the lmc , we merged the bica et al . and ogle catalogues .
we selected objects in the bica et al .
catalogues that were classified by them as `` c '' ( star cluster ) , `` ca '' ( cluster / association ) , `` ac '' ( association / cluster ) , or `` cn '' ( cluster with nebulosity ) . to determine ages and luminosities of the star clusters , we needed colors and magnitudes . for this , we had available to us images of the magellanic clouds taken by massey ( 2002 ) with the michigan curtis schmidt telescope at cerro tololo interamerican observatory ( ctio ) and a tektronix 2048@xmath112048 ccd .
massey obtained ubvr images and copious landolt ( 1992 ) standard stars .
he provided us with reduced ccd images and photometric calibrations to the standard johnson and cousins system .
the pixel scale of the ccd images is 2.32 .
although bvi photometry was available for the ogle catalog , we felt that the inclusion of u , the use of 4 filters , and the care with which massey calibrated his data made measuring photometry from the schmidt images worthwhile .
similarly , the fine compilations of integrated ubv photometry of clusters already in the literature ( van den bergh 1981 , bica et al .
1996 ) are for small subsets of our list of clusters and are not complete enough for our purposes .
for the lmc there are 11 fields each [email protected] , and for the smc there are 6 fields .
massey s ( 2002 ) fields together cover an area of 11.0 kpc@xmath1 of the lmc and 8.3 kpc@xmath1 of the smc .
thus , these fields cover most , although not all , of each galaxy .
( see massey s figures 1 and 2 for a sketch of the fields on large - field - of - view images of the smc and lmc ) .
we take these portions of the galaxies as representative of the galaxies as a whole and deal only with the cluster populations found on these images .
we used astrometric solutions of each v - band image to convert the ra and decs of the cluster catalogues to x , y pixel coordinates in each field .
we began by placing a 23 radius circle around each cluster position in each field and examining the region that was circled .
we found that in some cases there really was nothing in the circle that could be distinguished from the rest of the star field of the galaxy .
we discarded these clusters .
thus , to be included in our final analysis a cluster had to be visually distinguishable from the background galaxy and resolved with respect to an isolated star . in some cases , although there was a non - stellar object or concentration of stars in the aperture , we were dubious whether the object was really a star cluster .
usually these were tiny , faint objects that could be nothing more than a few stars superposed along the line of sight .
we flagged these objects as questionable , and they are plotted with different symbols and given lower weight in the analysis that follows .
the number of clusters in the lmc was 854 with 181 flagged as questionable . in the smc
there were 239 clusters with 106 flagged as questionable .
we also found that the cluster was not always in the center of the marked circle , and we adjusted the position accordingly . from visual examination , we also determined the size of the aperture needed for the photometry of the cluster .
the aperture was chosen to include the cluster , but exclude as much as possible extraneous field stars to the extent that this was obvious .
sky was determined from an annulus around the cluster with an interior radius that was 7 beyond the cluster aperture and 11.6 wide .
astrometry for each image was used to transform the position of each aperture from the v - band image to the other filters .
because the fields overlap , there were some clusters that appeared and were measured on more than one field . in the lmc
there were 247 duplicate measurements covering v from 9.5 to 16.5 and in the smc , 70 covering v from 10.5 to 15.5 .
a comparison of the photometry between pairs of measurements gives us an idea of the reliability of the photometry .
for both the lmc and smc , the average difference in the measured v magnitude was 0.16 magnitudes . for the smc the average differences in the colors was 0.04 , 0.05 , and 0.07 for u@xmath9b , b@xmath9v , and v@xmath9r , respectively .
for the lmc the average differences were 0.08 , 0.07 , and 0.08 .
for the brighter clusters with v less than 12.5 , the differences were 0.1 in v for both the lmc and smc , 0.03 and 0.04 for u@xmath9b for the two galaxies , 0.05 and 0.02 for b@xmath9v , and 0.03 for v@xmath9r .
thus , the photometry of the clusters appears to be reasonable .
we corrected the colors and magnitudes for reddening using a single reddening for each galaxy .
for the lmc we used an e(b@xmath9v ) of 0.13 mag . and for the smc 0.09 mag .
we used these with the extinction curve of cardelli , clayton , & mathis ( 1989 ) .
we used a distance modulus of 18.48 for the lmc and 18.94 for the smc .
motivated primarily by interest in the age distribution of clusters as a clue to the cluster formation history , people have long been interested in the ages of clusters in the magellanic clouds .
searle , wilkinson , & bagnuolo ( 1980 ) were perhaps the first to recognize that the integrated colors of star clusters in the magellanic clouds produce an age sequence on a color - color diagram ( ccd ) .
they used integrated uvgr colors to provide a relative ranking of the ages of 61 clusters in the large magellanic cloud .
frenk & fall ( 1982 ) then moved to the more common ubv - system in their study of 52 clusters in the lmc .
since then , various studies have worked on determining a transformation from ubv colors to age , including frenk & fall ( 1982 ) , elson & fall ( 1985 ) , chiosi , bertelli , & bressan ( 1988 ) , girardi & bica ( 1993 ) , and girardi et al .
others have also applied various techniques for determining ages of small samples , including examination of the upper asymptotic giant branch ( mould & aaronson 1982 ) , integrated spectroscopy ( rabin 1982 ) , and main - sequence photometry ( hodge 1983 ) .
others have examined the bigger clusters through the traditional method of placing individual stars on a color - magnitude diagram ( cmd ) and comparing to stellar evolutionary models .
the advantage of determining ages from integrated cluster colors is that the technique can be applied to large samples ; the disadvantage is that it is not as accurate .
the best method is cmd fitting of the individual stars within the cluster but this requires high resolution images in order to resolve the individual stars well enough for crowded - field photometric techniques to work and careful attention to the photometry and to subtraction of the interloper stellar population . because of concerns for these issues and because of the need for a complete sample of clusters , we have not used the ages determined by pietrzyski & udalski ( 1999 , 2000 ) from automatic photometry of the ogle project . instead
, we have used our ubvr integrated photometry and the new cluster evolutionary models that are now available . in this way we have a uniform way of determining ages for all clusters in our sample .
we recognize that the ages of individual clusters are uncertain , but expect that the sample taken as a whole will give an accurate picture of the cluster population .
specifically , we have used the leitherer et al .
( 1999 ) cluster models for the evolution to 1 gy , with the [email protected] models for the smc and the [email protected] models for the lmc .
we converted their v@xmath9r on the johnson system to the cousins system using the formula of bessell ( 1979 ) .
these are denoted ( v@xmath9r)@xmath13 . for ages beyond 1 gy , we used the ubv colors given by searle , sargent , & bagnuolo ( 1973 ) for 3 and 10 gy .
the searle et al .
colors at 1 gy match those given by the leitherer et al .
models , so the searle et al .
models join smoothly to the leitherer et al .
we then used the globular clusters for ( v@xmath9r)@xmath13 ( reed 1985 ) and the work by charlot & bruzual ( 1991 ) for the evolution of v with age for 110 gy .
we used two ccds for comparison of clusters and models : ubv and bvr .
the model tracks are different in these diagrams for the smc and lmc because of the difference in metallicity between the two galaxies .
one lmc field is shown in figure [ fig : lmcccd ] and one smc field is shown in figure [ fig : smcccd ] .
determining the age of each cluster is fraught with problems .
first , some clusters simply fall in parts of the diagram not visited by the models .
therefore , no age could be assigned and these clusters were eliminated . if a cluster was reasonable on one ccd and fell far from the models on the other , we assigned an age based on the diagram that was reasonable .
second , there are parts of the ccds where the models loop back on themselves so that clusters of different ages have similar colors .
this is particularly the case in the age range of about 630 my . for clusters falling in this region
, we did our best to determine a most likely age or took an average of possible ages .
a third source of uncertainty is stochastic effects due to small numbers of stars .
this affects the smaller clusters more than the rich clusters .
this problem is discussed and simulated by girardi & bica ( 1993 ; see also santos & frogel 1997 , brocato et al .
their simulations show a scatter of several tenths in ubv colors for small clusters as individual stars evolving within the cluster jerk the cluster colors around .
some of the richer clusters have well - determined ages based on main - sequence turn - offs or other features derived from stellar evolutionary models .
in addition , the brightest clusters were often saturated in some of our filters , making it impossible for us to determine some colors .
therefore , we adopted the ages in the literature that we felt were on solid ground . in the lmc
these clusters included ngc 1711 , ngc 1754 , ngc 1786 , ngc 1806 , ngc 1835 , ngc 1850 , ngc 1856 , ngc 1898 , ngc 1953 , ngc 2004 , sl 503 , and sl 569 ( hodge & lee 1984 , geisler et al .
1997 , olsen et al.1998 ) . in the smc three clusters had cmd ages : ngc 330 , ngc 361 , and ngc 416 ( carney et al .
1985 ; mighell , sarajedini , & french 1998a , b ) .
a comparison of our photometrically determined ages with the ages from cmds is given in figure [ fig : compareages ] .
we have plotted the clusters to show multiple ages determined from colors measured from different fields and to show the range in ages given from cmds .
ages determined from colors agree well with each other , but not always with that determined from the cmd .
there is no systematic trend except for the oldest clusters for which colors tend to underestimate the ages , giving ages of 13 gy where cmds give ages of 715 gy .
we have examined the uncertainties in the ages determined from integrated colors in another way .
we asked what age ranges would be consistent with colors extracted from the cluster evolutionary models at particular ages and given a typical uncertainty of @xmath140.05 mag .
the uncertainties in the log of the ages for the [email protected] and for the [email protected] models are given in table 1 .
the quantities that are tabulated are the absolute difference between the log of the extreme in the allowed ages and the log of the input age .
an uncertainty in the log of the age of order 0.100.15 dex is typical except in regions of the color - color diagrams where the cluster evolutionary tracks loop back on each other . for the [email protected] model
, the uncertainty goes up to 0.53 dex between 2060 my because of confusion with the 5 - 7 my part of the evolutionary track for 2030 my old clusters and the 1517 my part of the track for 4060 my old clusters . for clusters 7080 my there is confusion , within the allowed photometric uncertainties , with the 67 my and 1316 my parts of the track , and this causes the uncertainty to rise even higher to 1.06 dex . for the [email protected] model ,
confusion with the 1820 my part of the cluster evolutionary track causes an uncertainty of 0.31 dex in the log of the age for a 10 my - old cluster .
the 20 my - old cluster is confused with the 57 my and 912 my parts of the evolutionary tracks . for clusters with colors that have uncertainties greater than 0.05
mag , the uncertainties in the ages will , of course , be higher as well .
however , the preceding exercise assumed that clusters are evolving strictly along the cluster evolutionary tracks and only uncertainties in the colors are causing confusion .
the comparisons with ages determined from cmds given in figure [ fig : compareages ] may be more realistic and they paint a less optimistic picture .
there are not very many clusters with ages determined this way , but the cmds suggest that for clusters with ages @xmath15100 my the uncertainty in the log of the age is 0.5 dex , for 1001000 my it is 0.6 dex , and for @xmath161000 my it is 0.8 dex .
obviously , use of integrated colors is not the preferred way to determine the age of a cluster , but it is hoped that the ensemble of clusters will still be statistically representative for our purposes .
cluster evaporation could cause the color ages to be lower than the cmd ages because color ages are based on all the stars while cmd ages are based primarily on the most massive stars .
since low mass stars are preferentially lost during evaporation , clusters should have an excess of intermediate and high mass stars compared to the lowest mass stars , and therefore be slightly too blue compared to a model cluster with the full stellar initial mass function still present .
detailed modelling of the evaporation process will be needed to determine the conversion from color ages to cmd ages .
we note that the clusters with the most discrepant ages in our study are those closest to the evaporation limit discussed below . in the lmc we found that there are 5 star clusters with m@xmath8(10 my ) of @xmath17 to @xmath18 , implying that these are very massive star clusters .
all of these clusters are bright , with current m@xmath8 of @xmath19 to @xmath20 . however , four of them have ages 1415 gy .
these very old ages come from cmds and we can find no fault with the published work .
the fifth cluster has an age of 3 gy that comes from colors , but the colors look very reasonable , so we have no reason to discount the cluster .
therefore , we are forced to conclude that these really were very luminous when they were young , and that the lmc has hosted some extreme clusters in its distant past . in the smc
we also find 4 clusters with m@xmath8(10 my ) of @xmath21 to @xmath22 .
again , the clusters are bright and have ages of 28 gy .
the ages of two of these clusters come from cmds .
the other two come from the ccds but the colors fall on the models .
thus , it appears that even the smc has produced some extreme clusters in its distant past .
the final number of clusters in our sample for the lmc is 748 with 140 flagged as questionable from their appearance .
the total for the smc is 191 with 76 flagged as questionable .
the distribution of present - day absolute v magnitude of the clusters in the lmc is shown in figure [ fig : nmag ] .
the solid - line histogram in this figure and in the following figures are for the certain 608 clusters , while the dotted histograms are the additional 140 clusters that are questionable .
the distribution has a cutoff at the faint end with a magnitude limit of @xmath23 at the half - peak point , a long extension toward the bright end , and a number of very bright clusters that exceeds a power - law extrapolation at this bright end .
the questionable clusters are mostly near the faint end , as expected .
the various cluster ages are separated in figure [ fig : nmv ] .
the lower limit of brightness is the same for each age bin less than @xmath24 my , and the upper limit is about the same for each age too .
the oldest clusters are only massive clusters .
the absolute magnitude of a present - day cluster was converted to the absolute magnitude the cluster had at an age of 10 my using the colors and evolution models discussed in the previous section .
this value of m@xmath8(10 my ) was then converted to cluster mass using @xmath25 mag as the absolute magnitude of a 10 my old @xmath26 m_@xmath27 cluster with a metallicity of @xmath28 for the lmc ( leitherer et al .
1999 ) : @xmath29 figure [ fig : nmass ] shows the resulting distribution of cluster masses separated by age . the mass distribution shifts toward higher cluster mass with increasing age because of two effects .
first , the low mass end of the distribution shifts toward higher cluster mass because clusters fade with age , so higher mass clusters produce the same absolute magnitude at the limit of the survey as the age increases .
this shift in limiting detectable mass goes approximately as @xmath30 for @xmath31 , based on a power law fit to the tabulated m@xmath8(t ) in leitherer ( 1999 ) for metallicity @xmath28 ( see eqn . [
eq : limit ] and [ eq : mv ] below ) . thus the limiting mass increases by a factor of @xmath32 for each decade in age .
second , the upper limit of the cluster mass distribution increases with age because of the size - of - sample effect , in which larger numbers of clusters sample further into the high mass tail of the cluster mass distribution function .
the logarithmic time intervals in figure [ fig : nmass ] mean that older age intervals encompass longer time intervals and more cluster formation .
for a cluster mass function , the number of clusters as a function of the mass of the cluster , written in linear intervals of mass , @xmath33 the maximum likely cluster mass scales with the number @xmath34 of clusters as @xmath35 which comes from the equations @xmath36 and @xmath37 .
this correlation between @xmath34 and @xmath38 was observed directly by whitmore ( 2003 ) using a large number of galaxy surveys , and it was used by billett et al .
( 2002 ) and larsen ( 2002 ) to help explain the larsen & richtler ( 2000 ) correlation between the fraction of star formation in the form of clusters and the star formation rate . for @xmath4 in the likely range from 2 to 2.6 ,
the maximum mass scales with cluster number to a power between 1 and 0.62 .
figure [ fig : nmass ] plots cluster mass histograms in logarithmic intervals of time .
if the star formation rate is about constant , @xmath39constant , then the number of clusters that form in each log - time interval increases directly with time ( @xmath40 ) . thus the maximum mass increases with time @xmath41 in this context
, a constant star formation rate means a constant average rate , averaged over the time interval considered . if the star formation rate was larger at previous times , then the number of clusters formed in each log time interval increases faster than @xmath42 and the mass of the largest cluster in @xmath43 intervals increases faster than @xmath44 .
for example , if @xmath45 , then @xmath46 .
the similarity between the time dependence of @xmath47 and the time dependence of @xmath38 for @xmath48 to 2.5 explains why the mass distribution functions in figure [ fig : nmass ] shift to the right with increasing @xmath42 without changing their shape much .
an exception occurs for the oldest clusters , which are clearly lacking in low mass members although their highest masses fit the extrapolation from younger clusters . for these very old clusters
, evaporation is also depleting the lower masses , as we shall see momentarily .
the dashed lines in each histogram are reference lines with slopes of @xmath49 and @xmath50 on this log - log plot .
these correspond to @xmath51 and @xmath6 , which is in the range of solutions for our lmc data ( although this is not obvious from the mass functions in figure [ fig : nmass ] ) .
an interesting feature of figure [ fig : nmass ] is that the high mass ends of the distributions in the middle three age bins overlap in mass and are all more massive than the turnover masses , where fading becomes important for each time .
this means that the sum of the distributions , which is the total mass distribution for all clusters regardless of age , will also be a power law in this mass range , and the power will be about the same as it is in each age bin .
the distribution of cluster mass and age is shown in figure [ fig : agemass ] , following boutloukos & lamers ( 2003 ) .
the open circles are the questionable clusters .
the lower limit to the cluster mass , populated mostly by the questionable clusters , fits well to the fading limit ( solid line in figure [ fig : agemass ] ) , given by the equation @xmath52 where @xmath53 is the observed cutoff magnitude of the survey , from figure [ fig : nmag ] , and @xmath54 is the fading function from figure 47c in leitherer et al .
( 1999 ) . for ages larger than the limit in the leitherer et al .
table , which is 1 gy , we use a linear extrapolation of the tabulated values , which is @xmath55 for a @xmath56 m@xmath57 cluster .
equations [ eq : limit ] and [ eq : mv ] lead to equation [ eq : mfade ] above .
the dashed line in figure [ fig : agemass ] is the evaporation limit , given by the equation ( baumgardt & makino 2003 ) @xmath58 for cluster number @xmath59 , galactocentric radius @xmath60 , and galactic orbit speed @xmath61 in the lmc .
the two lines represent radii of @xmath62 kpc and 2 kpc , for which @xmath63 km s@xmath64 and 50 km s@xmath64 , respectively , from figure 6 in kim et al .
masses smaller than the limit given by the dashed line for their age should have evaporated by now .
the distribution of points in figure [ fig : agemass ] illustrates the simultaneous effects of fading and size - of - sample that were discussed above in reference to figure [ fig : nmass ] . the lower limit to the mass increases with time approximately as a power law @xmath65 from the fading limit , and the upper limit increases right along with it from the size - of - sample effect , keeping the total range in mass about constant for each time interval .
the maximum cluster masses for each logarithmic time interval are shown as plus - symbols .
these plus symbols are placed at the centers of the age intervals , so the corresponding dots for each symbol are shifted slightly to the left or right .
figure [ fig : maxmass ] shows the maximum cluster mass in each age interval , from the plus - signs in figure [ fig : agemass ] , versus the cluster age . in all cases
except one at a low age , the maximum mass cluster is a bonafide cluster , not a questionable cluster .
the open circles are for age intervals where the maximum mass cluster is close to or below the evaporation limit .
the distribution of points is fitted to a linear regression as @xmath66 .
this fit does not include the open circles because their masses could be severely depleted by evaporation .
a fit to ages less than 100 my gives a linear regression @xmath67 .
the maximum mass is expected to increase with the number of clusters , but the true number of clusters that ever formed in the lmc is not observed because of fading effects .
this figure corrects for this cluster loss by considering that they form at a constant rate , in which case the total number of clusters that formed in each log interval of time is proportional to the age . by plotting maximum mass versus age we can reconstruct the expected size - of - sample correlation without actually counting all the faded , destroyed , or evaporated clusters , which tend to be lower mass .
the slope of this correlation , @xmath68 below @xmath69 my and @xmath70 overall , suggests a cluster mass function with @xmath4 in the range from 1.95 to 2.35 , respectively , as given by equation [ eq : mmax ] .
if the highest mass clusters are underestimated in age , as suggested by figure [ fig : compareages ] , then the slope overall will be smaller than 2.35 .
thus we consider the initial cluster mass function in the lmc to have a slope in the range from 1.95 to 2.35 .
figure [ fig : maxmass ] can be used to find the cluster formation history in a galaxy if the initial cluster mass function ( icmf ) , destruction rate , and fading rate are known independently .
the icmf is the number of clusters as a function of the initial integrated mass of the cluster . if we consider that @xmath51 from the observation of young clusters , and we assume a cluster formation rate @xmath45 as above , then the overall slope of @xmath70 in the figure implies @xmath71 giving @xmath72 . in this case , the cluster formation rate in the age period from @xmath73 to @xmath74 my was smaller than that in the age period from @xmath75 to @xmath73 my by a factor of @xmath76 .
this assumes that the destruction time is longer than the fading time , as before , and that the color ages up to @xmath74 my are reasonably accurate .
figure [ fig : nmassall ] shows the total mass distribution function with all ages combined ( see also boutloukos & lamers 2003 ) .
this is the sum of the separate distributions in figure [ fig : nmass ] ; it is also a count of clusters projected against the ordinate in figure [ fig : agemass ] . for the lmc we have the fortunate situation where clusters more massive than the fading limit have a wide range of ages ( cf . fig .
[ fig : agemass ] ) , so the total mass distribution shows a power law section that is approximately the sum of the primeval power law sections from each separate age bin .
the flat part of the total distribution contains information about the cluster mass function as well .
both the flat and power - law parts of the total mass distribution were modelled using an initial cluster mass function , @xmath77 , that is a power law , and using the fading limit from equation [ eq : limit ] ( which considers the tabulation in leitherer et al .
( 1999 ) and our extrapolation to larger ages ) .
this model gives the expected number @xmath34 of clusters above the fading limit in each linear interval of mass as @xmath78 for adjustable constant @xmath79 in the case where @xmath80 for some maximum time of cluster formation , @xmath81 .
for @xmath82 , the number of clusters varies with mass as @xmath83 which is proportional to the initial cluster mass function , @xmath77 .
the plotted models are @xmath84 , which is appropriate for logarithmic intervals of mass .
this is @xmath85 for @xmath80 and @xmath86 for @xmath82 .
this expression for @xmath80 shows right away that the total mass distribution can have the observed flat part at low mass if @xmath87 .
the model assumes a nearly constant star formation rate ( as defined above ) and that fading alone provides the lower limit to the observed cluster mass .
this latter assumption is verified by the good fit between the fading line in figure [ fig : agemass ] and the lower limit to the observed cluster masses .
other destruction mechanisms are possible , such as collisions between clusters or between clusters and dense clouds .
our observation that fading contributed most to the lower mass limit implies that for all masses the fading time is less than the destruction time .
if there were a mass range where the destruction time was less than the fading time ( as might be the case in other galaxies see boutloukos & lamers 2003 ) , then we would have to take an upper limit to the integral in equation [ eq : n ] that is the minimum value of the fading time and the destruction time .
three maximum times are considered in figure [ fig : nmassall ] : 2 gy , 5 gy , and 7 gy , as indicated . for each time
we adjust the constant @xmath79 in front of equation [ eq : n ] to match the flat part of the observed @xmath88 distribution .
the excess low - mass clusters that come from the downward dips in @xmath89 at low @xmath42 in figure [ fig : agemass ] are ignored .
they do not affect the flat part of @xmath88 and tend to influence only the integral under the @xmath88 curve , which is the total number of clusters that forms . as a result ,
the theoretical fits for the next figure , which plots the number of clusters versus age , are slightly too high for the same parameters .
figure [ fig : nmassall ] has a kink at @xmath90 m@xmath57 and this is the mass corresponding to the fading limit at @xmath81 .
masses beyond this have not faded ( and apparently they have not been destroyed or evaporated either except at much larger mass cf .
[ fig : agemass ] ) , and so they have not been lost from our survey . as a result , the theoretical mass function integrated over all ages in figure [ fig : nmassall ] has about the same slope as the initial cluster mass function , @xmath77 , for @xmath91 m@xmath57 .
this slope is taken to have three values in the figure , ranging from 2 to 2.4 .
the model with initial cluster mass function slope @xmath51 is too steeply rising compared to the observations in the mass range from @xmath73 to @xmath92 m@xmath57 , but it is acceptable for the power law part above @xmath92 m@xmath57 , not counting the most massive clusters .
the case with @xmath93 fits the flat part best , it fits the power law part reasonably well if the most massive clusters are not included , and it has the best value for @xmath81 ( which was adjusted to match the observed kink in @xmath88 at @xmath92 m@xmath57 ) considering the observed distribution of ages in figure [ fig : agemass ] . the fading line in that figure was drawn with the extrapolation of the leitherer et al .
model out to @xmath94 gy , which was the best fit in figure [ fig : nmassall ] .
figure [ fig : nmassall ] shows an excess of massive clusters beyond @xmath95 m@xmath57 . where this excess begins determines the best fit to the power law part of the mass function ; if it starts at @xmath96 m@xmath57 , then the best fit is @xmath93 , but if it starts at @xmath97 m@xmath57 , then @xmath51 may be preferred .
the massive clusters are also the oldest clusters ( cf . fig .
[ fig : agemass ] ) . for these two reasons
, they appear to be a distinct population of clusters , like the halo globular clusters in the milky way .
their distributions do not appear to be extrapolations of the distributions for the less massive and younger clusters .
this allows for the possibility that the oldest massive clusters formed by a distinct mechanism , or that they formed by the same general mechanism but at a much higher rate than the disk clusters , with substantial evaporation and destruction of the lowest mass members , leaving only an excess of massive clusters now .
figure [ fig : nageall ] shows the distribution of cluster ages , summed over all masses ( see also boutloukos & lamers 2003 ) .
this is the count of clusters projected against the abscissa in figure [ fig : agemass ] .
the distribution is flat over times less than @xmath98 gy because the fading rate of clusters at the low mass end keeps pace with the broadening of the mass distribution at the high mass end , which comes from the size - of - sample effect at increasing @xmath99(t ) .
the three models are the same as in figure [ fig : nmassall ] .
the one with @xmath93 is the best fit to the flat slope of the distribution .
all of the models lie above the distribution because of the over - representation of low mass clusters at the wiggles in the @xmath100 function .
the theoretical models for figure [ fig : nageall ] come from the equation @xmath101 with the plotted quantity equal to @xmath102 in intervals of @xmath43 .
the constants @xmath79 for the three cases are the same as in figure [ fig : nmassall ] .
the flat part in this figure requires again @xmath103 . the age gap between @xmath104 gy and @xmath105 gy ( rich , shara , & zurek 2001 ) is evident from figure [ fig : nageall ] .
this gap is consistent with the age cutoffs we have assumed for the models and it reinforces our conclusion that the oldest massive clusters are a distinct population .
figure [ fig : agemass ] suggests that this gap is partly the result of severe evaporation , but the most massive clusters from this time period should still be visible if the star formation rate were the same as it is today .
as it is , the most massive clusters are close to the evaporation limit because the cluster formation rate was low .
there is no reason to think that the cluster formation rate was zero in this period , only that it was so low that the most massive likely cluster has suffered from evaporation . in that case , all the lower mass clusters would be gone or imperceptible by now .
returning to figure [ fig : maxmass ] , we see that the open circles from this age period have about the same maximum mass as those in the period from @xmath73 to @xmath74 my .
this means that the slope of a line drawn through these points is zero , and so @xmath106 giving @xmath107 independent of the icmf slope , @xmath4 .
thus the cluster formation rate was a factor of at most @xmath108 less in the period from @xmath98 to @xmath108 gy than it was in the period from @xmath73 - @xmath74 my .
the factor could have been smaller if the most massive clusters in this period lost mass by evaporation .
if our ages or masses turn out to be wrong for these few clusters , then the drop in the cluster formation rate could have been greater .
the clusters that are important for this time period are listed in table 2 .
analogous results for the small magellanic cloud are shown in figures [ fig : snmag ] to [ fig : snageall ] .
the faintness limit to the absolute magnitude for the smc is @xmath109 mag , from figure [ fig : snmag ] , and that determines the fading line in figure [ fig : sagemass ] . the fading model for the smc uses the z=0.004 case from figure 47d in leitherer et al . ( 1999 ) .
the smc distance is assumed to be 60 kpc . in this model ,
the absolute magnitude of a @xmath26 m@xmath57 cluster is @xmath110 for use in equation [ eq : mass ] , and the extrapolated fit in equation [ eq : mv ] is @xmath111 .
the rotation curve is from torres & carranza ( 1987 ) ; it suggests @xmath112 km s@xmath64 at @xmath62 kpc and @xmath113 km s@xmath64 at @xmath114 kpc for the evaporation limit in figure [ fig : sagemass ] .
the results for the smc are similar to those for the lmc : faintness limits the cluster masses at the low end and size - of - sample limits them at the high end .
the best fit to the cumulative mass and age functions and to the maximum mass - versus - age plot is for a icmf with a negative slope near @xmath93 for linear intervals of mass .
this gives the slope of @xmath115 in figure [ fig : smaxmass ] , which is the clearest measure of the icmf in this case .
the smc is also like the lmc in having old massive clusters that are not simple extrapolations of the total cluster mass function . figures
[ fig : snmass ] and [ fig : snmasall ] show these clusters well .
figure [ fig : sizemassboth ] shows the cluster masses versus sizes ( fwhm ) for both galaxies .
the symbol types indicate age .
the increase in mass with age shows up here as it did in earlier figures .
the young age symbols are at the bottom and the old age symbols are at the top .
there is no analogous shift of symbol - type to the right , which means that clusters do not get significantly larger with time for a given mass .
there is a slight trend for more massive clusters to be physically larger .
the maximum cluster size also increases with mass , as shown schematically with a line on the right of each panel that marks the outer envelope .
figure [ fig : agesizeboth ] shows the cluster size versus age .
there is a trend for the largest clusters to get larger with age , but this is probably the size - of - sample effect .
larger times correspond to larger time intervals on this log - abscissa plot , and to a larger total number of clusters forming in each log - time interval .
thus the maximum cluster mass increases to the right in the figure , as discussed before , and the maximum size increases toward the right also , along with the mass .
the line in the figure shows the predicted size - of - sample effect based on the relations between @xmath38 and time from figures [ fig : maxmass ] and [ fig : smaxmass ] and the schematic lines in figure [ fig : sizemassboth ] .
these relations are @xmath116 our size measurements are not as accurate as those of mackey & gilmore ( 2003a , b ) , who used hst data , but we have more clusters to see the size - of - sample effects better
. their correlation between cluster size and age , which is slightly steeper than ours , could have the same origin . in this case
, there would be no specific physical explanation for the occurrence of larger clusters at older ages , only a broader sampling of the massive end of the initial cluster mass function at these ages .
the agreement between the lower cluster mass limit as a function of age in figures [ fig : agemass ] and [ fig : sagemass ] and the fading limit suggests that clusters are lost from view mostly by fading and not by destruction in these two galaxies .
this is consistent with the conclusion by boutloukos & lamers ( 2003 ) who found a long destruction time for the smc ( @xmath117 yrs ) .
they did not analyze data for the lmc .
our overall result for the smc differs from theirs , however , because we do not attempt to fit two distinct power law parts to the cluster mass distribution integrated over age or to the cluster age distribution integrated over mass , but instead we fit both distributions as a whole to the fading - statistical model .
boutloukos & lamers also assumed an icmf slope of @xmath118 for the smc , but the flat part of the age distribution integrated over mass requires a steeper initial function , @xmath93 , unless the star formation rate was smaller by a factor of @xmath119 from @xmath73 to @xmath74 my ago than it is today .
figure [ fig : maxmass ] suggested a new way to determine the icmf from the size of sample effect . size - of - sample effects appear as a hidden influence in many studies of star clusters .
they contribute to the impression that starburst regions form more massive clusters ( billett et al .
2002 ; larsen 2002 ) .
they may also account for the increase in cluster size with age that was found by mackey & gilmore ( 2003a , b ) , as shown in the previous section . in figure
[ fig : maxmass ] , the size - of - sample effect appears as a correlation between the age and the mass of the most massive cluster at that age ( in logarithmic age bins ) .
this may be a better way to determine the icmf than the slope on a log - log histogram of cluster mass for either a single age ( which can have poor statistics ) or integrated over age ( which can give the wrong slope because of the age mixture ) .
the icmf slopes found here are in the range from @xmath118 to @xmath120 , depending on the cluster formation history .
the steeper end of this range exceeds the slope of @xmath118 found in the antenna galaxy ( zhang & fall 1999 ) , but is consistent with values found by larsen ( 2002 ) for several galaxies after he corrected for size - of - sample effects .
the oldest clusters in the lmc and smc are all massive .
lower mass counterparts could have been observed above the fading limit , but they are not present .
they may have evaporated because the oldest clusters are close to the evaporation limit for all masses in figures [ fig : agemass ] and [ fig : sagemass ] .
alternatively , the oldest clusters in the lmc and smc could have had a different initial mass distribution , such as a gaussian distribution in log - mass ( vesperini 2001 ) .
we can not tell the difference between this initial function and a power law because the number of massive clusters is too small and the evaporation mass limit is too high ( all lower mass clusters have evaporated ) .
de grijs , bastian , & lamers ( 2003 ) suggest that an initially power - law cluster mass function in m82 has begun to turn over at low mass in its gradual progression toward a gaussian .
m82 is a better place to observe this than the lmc or smc because the number of clusters and the tidal density in m82 are both large .
the large tidal density makes the cluster evaporation time smaller , and then fading is less severe for clusters in the last stages of evaporation .
the integrated properties of clusters in the lmc and smc were measured from ground - based images and analyzed to determine the likely slope for the power - law form of the initial cluster mass function .
this slope can not be determined by conventional means because the blend of ages prevents the combined cluster mass function from revealing the initial function , and because the sample is too small to get a statistically significant cluster mass function in a narrow age range .
we could still determine the initial cluster mass function from the data , however , using a combination of theoretical fading limits and the size - of - sample effect .
the size - of - sample effect follows from the assumption that clusters randomly sample a physically - determined mass function and so larger samples are likely to have larger most - massive clusters .
the results suggest that the initial cluster mass function is a power law with a slope between @xmath118 and @xmath120 for the lmc and smc , nearly independent of time .
the slope was determined independently using three methods : a fit of the model to the total mass function integrated over all cluster ages , a fit of the same model to the number of clusters versus age integrated over all masses , and a least - squares fit to the maximum cluster mass versus age .
if there is a significant number of high mass clusters that are not part of the icmf , then @xmath93 is preferred for each galaxy . the maximum cluster mass and the maximum cluster size increase with age when plotted in equal intervals of log - age because the total number of clusters increases linearly with age in such a distribution .
then , as the number of clusters increases , the high ends of the distribution functions get sampled further out . the lower limit to the observed cluster mass increases with age too because of fading .
both vary with age in about the same way for the lmc and smc , so the histogram of cluster mass shifts in a nearly self - similar way toward higher cluster masses for aging clusters .
the oldest clusters in the lmc and smc stand apart from the most obvious extrapolations found for younger clusters .
they are all very massive and they lie far above the fading limits . either a large population of them has already evaporated or low mass clusters did not form in early times .
we are grateful to dr .
philip massey for the use of his schmidt images and calibrations .
tjd would like to thank the national science foundation for funding the research experience for undergraduates program at northern arizona university under grant 9988007 to northern arizona university and dr .
steven tegler for running the program .
mm acknowledges the mit field camp at lowell observatory run by dr .
jim elliot .
support to dah for this research came from the lowell research fund and grant ast-0204922 from the national science foundation .
support to bge came from grant ast-0205097 from the national science foundation .
baumgardt , h. , & makino , j. 2003 , mnras , 340 , 227 bessel , m. s. 1979 , pasp , 91 , 589 bica , e. l. d. , clari , j. j. , dottori , h. , santos , j. f.c . , & piatti , a. e. 1996 , apjs , 102 , 57 bica , e. l. d. , & dutra , c. m. 2000 , aj , 119 , 1214 bica , e. , l. d. , schmitt , h. r. , dutra , c. m. , & oliveira , h. l. 1999 , aj , 117 , 238 billett , o. h. , hunter , d. a. , & elmegreen , b. g.2002 , aj , 123 1454 boutloukos , s. g. , & lamers , h. j. g. l. m. 2003 , mnras , 338 , 717 brocato , e. , castellani , v. , raimondo , g. , & romaniello , m. 1999 , a&as , 136 , 65 cardelli , j. a. , clayton , g. c. , & mathis , j. s. 1989 , apj , 345 , 245 carney , b. w. , janes , k. a. , & flower , p. j. 1985 , aj , 90 , 1196 charlot , s. , & bruzual , a. g. 1991 , apj , 367 , 126 chiosi , c. , bertelli , g. , & bressan , a. 1988 , a&a , 196 , 84 de grijs , r. , bastian , n. , & lamers , h. j. g. l. m. 2003 , apj , 583 , l17 elson , r. a. w. , & fall , s. m. 1985 , apj , 299 , 211 frenk , c. s. , & fall , s. m. 1982 , mnras , 199 , 565 geisler , d. , bica , eduardo , dottori , h. , clara , j. j. , piatti , a. e. , & santos , j. f. c. , jr .
1997 , aj , 114 , 1920 gelatt , a. e. , hunter , d. a. , & gallagher , j. s.2001 , pasp , 113 , 142 girardi , l. , & bica , e. 1993 , a&a , 274 , 279 girardi , l. , chiosi , c. , bertelli , g. , & bressan , a.1995 , a&a , 298 , 87 hodge , p. w. 1961 , apj , 133 , 413 hodge , p. w. 1983 , apj , 264 , 470 hodge , p. w. 1986 , pasp , 98 , 1113 hodge , p. w. 1988 , pasp , 100 , 1051 hodge , p. w. , & lee , s .- o .
1984 , apj , 276 , 509 kim , s. , staveley - smith , l. , dopita , m. a. , freeman , k. c. , sault , r. j. , kesteven , m. j. , mcconnell , d. 1998 , apj , 503 , 674 kndlseder , j. 2000 , a&a , 360 , 539 kontizas , e. , metaxa , m. , & kontizas , m. 1988 , aj , 96 , 1625 kontizas , m. , morgan , d. h. , hatzidimitriou , d. , & kontizas , e. 1990 , a&as , 84 , 527 landolt , a. u. 1992 , aj , 104 , 340 larsen , s. s. 2002 , aj , 124 , 1393 larsen , s. s. , & richtler , t. 2000 , a&a , 354 , 836 leitherer , c. , schaerer , d. , goldader , j. d. , gonzlez delgado , r. m. , robert , c. , kune , d. f. , de mello , d. f. , devost , d. , & heckman , t. m. 1999 , apjs , 123 , 3 mackey , a.d .
, & gilmore , g.f .
2003a , mnras , 338 , 85 mackey , a.d . , & gilmore , g.f .
2003b , mnras , 338 , 120 massey , p. 2002
, apjs , 141 , 81 mighell , k. j. , sarajedini , a. , & french , r. s. 1998a , aj , 116 , 2395 mighell , k. j. , sarajedini , a. , & french , r. s.1998b , apj , 494 , l189 mould , j. , & aaronson , m. 1982 , apj , 263 , 629 olsen , k. a. g. , hodge , p. w. , mateo , m. , olszewski , e. w. , schommer , r. a. , suntzeff , n. b. , & walker , a. r. 1998 , 300 , 665 pietrzyski , g. , & udalski , a. 1999 , aca , 49 , 157 pietrzyski , g. , & udalski , a. 2000 , aca , 50 337 pietrzyski , g. , udalski , a. , kubiak , m. , szymaski , m. , woniak , p. , & ebru , k. 1999 , aca , 49 , 521 pietrzyski , g. , udalski , a. , kubiak , m. , szymaski , m. , woniak , p. , & ebru , k. 1998 , aca , 48 , 175 rabin , d. 1982 , apj , 261 , 85 reed , b. c. 1985 , pasp , 97 , 120 rich , r.m . , shara , m.m . , &
zurek , d. 2001 , aj , 122 , 842 santos , j. f. c. , jr .
, & frogel , j.a .
1997 , apj , 479 , 764 searle , l. , sargent , w. l. w. , & bagnuolo , w. g. 1973 , apj , 179 , 427 searle , l. , wilkinson , a. , & bagnuolo , w. g. 1980 , apj , 239 , 803 torres , g. , & carranza , g.j .
1987 , mnras , 226 , 513 van den bergh , s. 1971 , a&a , 12 , 474 van den bergh , s. 1981 , a&as , 46 79 vesperini , e. 2001 , mnras , 322 , 247 whitmore , b. c. 2003 , in stsci symp .
14 , a decade of hst science , ed .
m. livio , k. noll , & m. stiavelli ( baltimore : stsci ) zhang , q. , & fall , s.m .
1999 , apj , 527 , l81 lllccc ogle - lmc0531 & 5:30:02.05 & -69:31:36.14 & 5 & @xmath121 & c + kmk88 - 38.h88 - 206 & 5:12:09.30 & -68:54:40.58 & 6.5 & @xmath122 & c + ogle - lmc0169 & 5:10:06.07 & -69:05:19.66 & 10 & @xmath123 & c + bsdl917 & 5:13:04.00 & -70:26:55.00 & 10 & @xmath123 & c + kmhk898 & 5:26:23.95 & -68:02:48.09 & 10 & @xmath124 & q + | the properties of @xmath0 star clusters in the large and small magellanic clouds were determined from ground - based ccd images in ubvr passbands .
the areal coverage was extensive , corresponding to 11.0 kpc@xmath1 in the lmc and 8.3 kpc@xmath1 in the smc .
after corrections for reddening , the colors and magnitudes of the clusters were converted to ages and masses , and the resulting mass distributions were searched for the effects of fading , evaporation , and size - of - sample bias .
the data show a clear signature of cluster fading below the detection threshold .
the initial cluster mass function ( icmf ) was determined by fitting the mass and age distributions with cluster population models .
these models suggest a new method to determine the icmf that is nearly independent of fading or disruption and is based on the slope of a correlation between age and the maximum cluster mass in equally spaced intervals of log - age . for a nearly uniform star formation rate , this correlation has a slope equal to @xmath2 for an icmf of @xmath3 .
we determine that @xmath4 is between @xmath5 and @xmath6 for the lmc and smc using this method plus another method in which models are fit to the mass distribution integrated over age and to the age distribution integrated over mass .
the maximum mass method also suggests that the cluster formation rate in the lmc age gap between 3 and 13 gy is about a factor of ten below that in the period from 0.1 gy to 1 gy .
the oldest clusters correspond in age and mass to halo globular clusters in the milky way .
they do not fit the trends for lower - mass clusters but appear to be a separate population that either had a very high star formation rate and became depleted by evaporation or formed with only high masses . |
far from the energizing regions of star formation in the milky way and other galaxies , the interstellar medium ( ism ) is still roiling with activity , and rife with structure . both the microphysical properties and turbulent activity have been increasingly well characterized by galactic and extragalactic radio observations . in particular , recent high - resolution galactic emission surveys in the 21 cm hydrogen line ( e.g. @xcite ) , combined with galactic absorption surveys ( e.g. @xcite ) , and mapping of face - on external galaxies ( e.g. @xcite ) , have begun to provide a wealth of thermal and kinematic information about the atomic ism component , which comprises the majority of the total ism mass in most spiral galaxies .
analysis of this data promises to yield a detailed empirical description of the atomic gas , which is known to consist of both warm and cold components , and to be strongly turbulent ( e.g. @xcite ) . as observations of the ism advance ,
there is a need on the theoretical side for increasingly sophisticated ism modeling . with modern computational tools
, it is possible to pursue time - dependent hydrodynamic models which incorporate many physical processes .
this numerical modeling can extend established `` classical '' results for simplified systems into more realistic regimes , and test conceptual proposals for the behavior of complex systems in a rigorous fashion .
the goal of detailed ism modeling , of course , is not sophistication for its own sake , but to gain understanding about how different `` elemental '' processes interact , to ascertain which among many contributing processes are most important , and to aid in interpreting and developing reliable physical diagnostics from observations . broadly , the presence of structure in the atomic ism can be easily understood as a consequence of the bistable thermal equilibrium curve over a range of pressures , including those typical of the ism .
since the temperatures of the two stable thermal equilibria differ by a factor of @xmath1 ( at fixed pressure ) , the `` classical '' expectation based on the principle of pressure equilibrium is a system of cold , dense clouds embedded in a much more diffuse warm intercloud medium @xcite . thermal instability ( ti )
tends to move gas parcels at intermediate temperatures into one of the stable phases @xcite .
clouds are initially expected to condense at preferred scales where conduction limits local thermal gradients .
while these basic processes are certainly involved in establishing the ism s structure , the end result is a complex product of evolution and interactions with other physical processes , leaving many open questions .
for example , how do the agglomerations and disruptions of cold clouds depend on the turbulence properties , and how does this affect the mass function of condensations that results ?
many processes have been proposed that can produce turbulence in the ism ( see e.g. @xcite for recent reviews ) .
traditionally , turbulence is thought to be driven primarily by supernovae @xcite ( and , to a lesser extent , expanding hii regions ) , because the total kinetic energy they are able to supply could be sufficient to offset the turbulent dissipation in the milky way s diffuse ism ( spitzer 1978 , ch .
supernovae are certainly the primary source of turbulence near regions of high - mass star formation .
however , it is not clear how effectively this energy can in fact be shared with the bulk of the ism , so other sources may be ( or combine to be ) of comparable importance .
indeed , observations indicate that the levels of turbulence are not strongly correlated with spiral arms ( where star formation is enhanced ) , and are just as large in outer galaxies ( where overall star formation rates are low ) as in inner regions @xcite
. moreover , recent 3d simulations @xcite in which turbulence is driven solely by supernovae find that velocity dispersions are significantly lower in cold gas than in warm gas , inconsistent with observations @xcite .
an obvious non - stellar energy source for the ism is galactic rotation . wherever the angular velocity decreases outward and magnetic fields are present ,
the powerful magnetorotational instability ( mri ) is expected to tap this rotation and drive large - amplitude ism turbulence @xcite .
detailed development of mri has primarily been studied in adiabatic or isothermal gas , where turbulent velocities and alfvn speeds grow into rough equipartition at slightly subsonic levels ( e.g. @xcite hereafter hgb1 , hgb2 ) ) .
adiabatic and isothermal models , however , are essentially single phase , with only small variations in density and temperature .
how do turbulent saturation levels differ in a medium where there are huge variations in conditions , such that subsonic speeds with respect to the diffuse gas are highly supersonic with respect to the dense gas ? in the real ism , dynamics must affect thermodynamics , and vice versa .
the turbulent power input is significant , and both ( irreversible ) dissipative heating and ( reversible ) pdv heating and cooling can alter distributions of temperatures compared to the narrow spikes at warm and cold equilibria that would otherwise occur . in turn
, thermodynamics potentially can affect loss rates of turbulence : supersonic compressions are dissipative while subsonic compressions are not , and dissipation of magnetic energy by reconnection depends on local conditions of density and temperature .
cloudy structure also changes effective flow `` collision '' times , as well as field line geometry .
indeed , recent observational evidence has shown that the fraction of unstable gas in the ism may be significant ; @xcite found that at high latitudes , about half the warm neutral medium ( wnm ) lies at thermally unstable temperatures between 500 - 5000 k. numerical models which include effects of star formation @xcite find both turbulence and significant fractions of unstable gas , although it is not clear how much the temperature distributions are affected by the direct heat inputs in the star formation feedback algorithms of these models .
recent simulations have addressed nonlinear evolution , in 2d and 3d , of ti in the ism without `` stellar '' energy inputs @xcite , and there have also been many numerical studies , in 2d and 3d , of the mri in single - phase gas . in previous work
, we performed 2d studies of ti and mri in combination ( @xcite , hereafter paper i ) .
paper i showed that mri growth rates in a two - phase medium are comparable to those in a single - phase medium with the same @xmath9 and @xmath10 , provided that the cloud separation along field lines does not exceed half of the fastest - growing mri wavelength ( typically @xmath11 ) .
although there have been suggestions that ti itself could be a significant source of turbulence , `` pure ti '' models we performed show that for pressures comparable to mean galactic values ( i.e. away from hii regions or recent supernovae ) , velocity dispersions are only a few tenths of a @xmath12 . in our 2d simulations ,
the mri leads to large - amplitude velocities and magnetic fields , but as for single - phase 2d models , late time behavior is dominated by the `` channel flow ; '' quasi - steady turbulence is possible only for 3d flows .
the present work constitutes the extension of paper i to 3d , in order to study the saturated state of mri in the presence of a two - phase medium .
as we shall describe , we have performed a variety of simulations , with parameters covering a range of conditions characteristic of the atomic ism .
the plan of this paper is as follows : in 2 we briefly describe the numerical method , and the initializations for the various models we have performed . in
3 we present the results of our simulations in terms of the models physical structure , thermodynamic distributions , and turbulent states ( in velocities and magnetic fields ) , as well as exhibiting sample synthetic observations based on our simulated data .
we summarize , discuss the astronomical implications of our results , and compare to previous work in 4 .
the numerical methods utilized for the present study are essentially the same as those of paper i , but extended from 2d to 3d . for a complete description of the numerical method and tests , please see that work . here , we briefly summarize the salient points .
we integrate the time - dependent equations of magnetohydrodynamics with a version of the zeus code @xcite .
zeus uses a time - explicit , operator - split , finite difference method for solving the mhd equations on a staggered mesh , capturing shocks via an artificial viscosity .
velocities and magnetic field vectors are face - centered , while energy and mass density are volume - centered .
zeus employs the ct and moc algorithms @xcite to maintain @xmath13 and ensure accurate propagation of alfvn waves .
we have implemented volumetric heating and cooling terms , and a thermal conduction term .
the update due to net cooling is solved implicitly using newton - raphson iteration . for a given hydrodynamical time step ,
the change in temperature in each zone is limited to be less than 25% .
this is a somewhat larger fraction than the 10% limit used in paper i , which allows us to run with larger time steps needed to make 3d calculations practical .
tests have shown that relaxing this constraint does not affect cloud structure ; @xmath14 t exceeds 10% only in a very small fraction of zones .
the conduction term is solved explicitly using a seven point stencil for the second derivative of temperature .
we also model the differential rotation of the background flow and the variation of the stellar / dark matter gravitational potential in the local limit with @xmath15 , where @xmath16 is the galactocentric radius of the center of our computational domain .
the equations we solve are therefore : @xmath17 @xmath18 @xmath19 @xmath20 all symbols have their usual meanings .
the net cooling per unit mass is given by @xmath21 .
we adopt the simple atomic ism heating and cooling prescriptions of @xcite , in which the cooling function , @xmath22 , is a piecewise power - law fit to the detailed models of @xcite .
the heating rate , @xmath23 , is taken to be constant at 0.015 @xmath24 . in the tidal potential term of equation ( [ mom ] ) , @xmath25 is the local dimensionless shear parameter .
we adopt @xmath26 equal to unity , to model a flat rotation curve in which the angular velocity @xmath27 .
the present set of simulations is 3d , with the computational domain representing a cubic sector of the ism in the radial - azimuthal - vertical ( @xmath28 ) domain .
we employ shearing - periodic boundary conditions in the @xmath29-direction @xcite , and periodic boundary conditions in the @xmath30- and @xmath31-directions , as originally implemented in zeus by @xcite .
this framework allows us to incorporate realistic galactic shear , while avoiding numerical artifacts associated with simpler boundary conditions .
we have previously used an isothermal version of the same code to study larger - scale galactic ism problems @xcite .
we have parallelized the code with mpi to run on distributed - memory platforms .
we perform a standard domain decomposition in @xmath29 , so that each processor works on a slab of the domain .
decomposing in @xmath30 and @xmath31 as well would reduce message - passing time when running on more than eight processors , but for gigabit networks and faster the compute time exceeds the message - passing time for 128 or fewer processors in any case .
thus , for our moderate problem sizes , the additional effort that would be required to parallelize the shearing - periodic boundary conditions is not merited .
our standard resolution is @xmath32 zones . to confirm numerical convergence
, we also performed one simulation at @xmath33 and found the results to be similar to the standard resolution run .
our box is 200pc on each side , giving a resolution of about @xmath34 in our @xmath33 run , and about @xmath35 in our @xmath32 resolution runs .
we set the conduction coefficient to @xmath36 .
this level of conduction was chosen to allow us to resolve all modes of ti that are present ( see paper i ) . in our fiducial model ,
the initial state of the gas is constant density , constant pressure with conditions comparable to mean values in the solar neighborhood ; @xmath37 and @xmath38 .
the corresponding initial isothermal sound speed is @xmath39 .
the initial magnetic field strength satisfies @xmath40 , corresponding to @xmath41 g in the vertical ( @xmath31 ) direction .
this is our `` standard '' run .
additional simulations are performed with varying mean densities of @xmath42 and 4.0 @xmath43 , as well as one with lower magnetic field strength , @xmath44 .
we also performed an isothermal simulation with @xmath45 and @xmath46 .
this value of @xmath47 was chosen so that the initial thermal pressure matches the mean late - time pressure in our cooling models .
finally , we also performed a simulation with heating and cooling turned on that was initialized from the saturated - state , turbulent isothermal model . for all our models we adopt the galactic orbital period at the solar radius , @xmath48 , to normalize the shear rate .
since increasing or decreasing the mean density by a large factor relative to @xmath49 would initialize the gas in a thermally @xmath50 state , some of our simulations are initialized with a medium already in a two - phase state , rather than with a uniform density . for these models ,
spherical clouds of cold dense gas are inserted into a warm ambient medium at random locations .
the number of clouds is adjusted so that the average density of the cloudy medium is at the desired level .
a similar simulation was performed in paper i , which allowed us to study the growth rates of the mri in an initially quiescent cloudy medium . since the 2d simulations of paper i were axisymmetric there was no evolution of the model until mri modes began to grow .
this allowed us to compare directly the mri growth rates of an adiabatic run with a two - phase run , illustrating the effect of cloud size and distribution on the growth rates .
in the present 3d simulations , however , the evolution is rapid because the symmetry in the azimuthal direction is broken .
individual clouds are sheared out relatively quickly , and also begin to merge with nearby clouds .
nevertheless , because mri - driven turbulence eventually dominates both the initially - thermally - unstable and initially - two - phase models , at late times the two are indistinguishable . on top of the initial conditions given above , we add pressure perturbations with a white noise spectrum at the 0.1% level to seed the ti and mri . in the next section ,
we describe results from our standard run in detail , and comment on differences with the other runs as is appropriate .
figures [ vol1 ] and [ vol2 ] are volume renderings of the 3d density data cube , from our run with fiducial parameters , and resolution @xmath33 , at @xmath51 and @xmath52 orbits .
the early development of both ti and mri in the present set of 3d simulations is quite similar to the development previously described for 2d simulations in paper i. initially the gas is thermally unstable .
the cooling time scale is much shorter than the orbital time scale , and the gas quickly separates into many small , cold clouds embedded in a warm ambient medium .
this phase of the evolution lasts about 20 myr , which is comparable to the 2d simulations of paper i. the typical size scale of the clouds is about 5 pc , consistent with expectations for the fastest growing modes at the adopted level of conductivity .
the size scale of the clouds is still fairly close to its initial distribution in figure [ vol1 ] at @xmath51 orbits .
after the initial condensation phase of ti is complete , large scale galactic shear begins to drive the evolution . already at @xmath53 orbits ,
the clouds have become elongated in the @xmath30 direction . during the first few orbits interactions take place between nearby clouds , which typically lead to mergers , increasing the typical size scale significantly . at about @xmath54 orbits ( @xmath55 )
the modes of the mri have grown significantly and now begin to dominate the evolution of the model .
the simulation becomes fully turbulent , drastically altering the dynamics compared to the axisymmetric model of paper i. shear from the mri with velocities in all directions , combined with galactic shear with velocities in the azimuthal direction , leads to repeated disruptive interactions and collisions between clouds .
clouds merge into an interconnected network , with individual entities existing for only short periods of time .
it is difficult to convey the dynamical nature of the simulations to the reader using only snapshots in time ; the animation associated with figures [ vol1]-[vol2 ] shows this much more clearly .
while the structure remains highly dynamic , a quasi - equilibrium saturated state is established by @xmath56 orbits , and the statistical properties of the gas remain relatively constant throughout the latter half of the simulation ( up to t=10 orbits ) .
the approach to a quasi - steady turbulent state in these models is generally similar to the results for isothermal or adiabatic single phase models ( e.g. hgb1 , hgb2 ) . in the remainder of 3
, we discuss details of evolution and quasi - steady properties , similarities and differences from single - phase models , and dependencies on model parameters . the density probability distribution functions ( pdfs ) from our standard run ( at @xmath32 )
are shown in figure [ denpdf ] at @xmath57 and @xmath58 orbits .
we show both mass - weighted and volume - weighted density pdfs in panels a - d , and compare the pdfs of the @xmath32 and @xmath33 runs in panel d. similar to our results in paper i , we find that by mass , most of the gas is in the cold phase , while the warm phase occupies most of the volume . after the initial development of ti has completed , at @xmath51 orbits , the mass fraction of gas in the warm ( f ) , unstable ( g ) , and cold ( h ) phases is 14% , 5% , and 80% , respectively . by volume ,
83% , 9% , and 8% of the gas resides in the warm , unstable , and cold phases . from @xmath51 to @xmath59 orbits ( panels a and b of figure [ denpdf ] ) the evolution is driven mainly by galactic shear .
the size distribution of the clouds shifts to larger masses through mergers , but the density pdfs over this interval vary little .
the fraction of gas in each phase changes by only a few percent during this time period .
in contrast , between @xmath59 and @xmath60 , panels b and c of figure [ denpdf ] , the evolution changes from being driven primarily by galactic shear , to being driven primarily by the mri .
the model becomes fully turbulent , and this has a significant effect on the detailed shape of the density pdf .
the fractions of gas in the warm , unstable , and cold phases at @xmath60 are now 10% , 7% , and 83% by mass , and 84% , 8% , and 7% by volume . near the end of the simulation , at @xmath61 , the gas fractions are 14% , 18% , and 67% percent by mass and 82% , 10% , and 6% percent by volume . from @xmath62 to @xmath61 , ( panel d of figure [ denpdf ] )
the pdf remains very similar , indicating that the model has reached a quasi - steady state . at late times
, gas is found at both lower and higher densities than was previously observed before the development of the mri .
thus , the magnetized turbulence induces both strong compressions and significant rarefactions . compared to the maximum ( @xmath63 ) and minimum ( @xmath64 ) densities before the onset of turbulence , @xmath63 increases by an order of magnitude and @xmath64 decreases by a factor of about 3 .
the fraction of gas in the intermediate density regime is a factor 2 3 larger after the full development of mri compared to early on .
the proportion of thermally - unstable gas is never greater than 20% of the whole ( for this set of parameters ) , but exceeds the proportion of thermally - stable warm gas during the turbulent stages of evolution . to investigate properties of individual condensations in our model
, we use an algorithm similar to that of clumpfind @xcite .
the algorithm was developed and applied by @xcite to identify clumps in simulations of turbulent molecular clouds .
briefly , the algorithm first finds all local maximum values of density in the computational volume .
all grid cells with a density higher than a chosen threshold density , @xmath65 , are assigned to the nearest local maximum .
this set of continuous zones defines a clump .
the only other parameter needed is a smoothing length , applied to the initial density data cube ( see gammie et al .
2003 ) ; we set this to 1.5 grid zones . in figure [ mass_dn ]
we show the clump mass spectrum for two different choices of threshold density , @xmath66 and 20 @xmath43 .
this mass spectrum is computed at @xmath67 orbits .
mass spectra from other late times are similar . with @xmath68 ,
812 clumps were found , with a minimum clump mass of @xmath69 , and a maximum mass of @xmath70 . for reference ,
the total mass in the simulation is @xmath71 .
increasing the critical density to @xmath72 , we find 168 clumps , with a minimum mass of @xmath73 , and a maximum mass of @xmath74 .
for both cases , the peak of the mass spectrum is in the range @xmath75 ; the peak increases slightly for larger @xmath65 . to describe their shapes ,
we compute diagonalized moment of inertia tensors for each clump , following @xcite .
figure [ shape ] plots the ratios , for each clump , of the smallest ( c ) and intermediate ( b ) axes to the largest ( a ) axis .
prolate - shaped clumps lie near the diagonal line , oblate clumps lie near the right side vertical axis , and triaxial clumps lie in the center . using two dotted lines to demarcate these groups
, we find 38 % of the clumps are prolate , 49 % are triaxial , and 14% are oblate .
although clumps are certainly not round , typical minimum to maximum axis ratios are about 2:1 .
`` filaments '' , with c / a=0.1 are common , however , and these elongated structures are easy to pick out in figure [ vol2 ] .
the pressure pdfs at @xmath76 and 9 orbits are presented in figure [ presspdf ] . at @xmath77 ,
most of the gas falls within a narrow range of pressures , p / k=900 - 1300 @xmath78 .
this is lower than p / k=2000 @xmath78 in the initial conditions , due to systematic cooling in the thermally unstable stage of evolution .
the pressure pdf changes little from @xmath77 to @xmath79 orbits , shown in panels a and b of figure [ presspdf ] . with the development of mri ,
however , gas is driven to both higher and lower pressures , as can be seen in panels c and d , at @xmath62 to @xmath61 orbits .
the mean volume - weighted pressure at the end of the simulation is slightly lower than that after ti has developed , about p / k=1200 @xmath78 .
the pressures in the cold and warm phases are approximately equal in the latter half of the simulation , while the pressure in the intermediate phase is slightly higher , about p / k=1300 @xmath78 .
the dispersion in pressure early in the simulation is about @xmath80 , while late in the simulation this increases to as much as @xmath81 . in figure [ coolcurve ]
we show scatter plots of pressure against density overlayed on our model cooling curve at @xmath76 and 9 orbits .
we also show contours of constant temperature to indicate the transitions between different phases of gas .
only a fraction of the zones are included because of the large number of cells contained in our 3d simulations .
early in the simulation ( panels a and b ) , the gas is close to pressure equilibrium , although high density gas lies closer to the thermal equilibrium curve .
later in the simulation ( panels c and d ) , strong interactions between clouds can drive gas far from pressure equilibrium . at low densities where the cooling time scale is longer than the dynamical time
, gas can be found at pressures as high as p / k=3200 @xmath78 and as low as 800 @xmath78 , a range of a factor four .
much of the low - density gas is not in thermal equilibrium . in high density regions
there is also a wide range of pressures observed ( p / k=800 - 4000 @xmath82 ) , but because the cooling time is very short ( @xmath83 yr ) this gas maintains thermal equilibrium . at early times , distributions of density and pressure
are quite similar to the corresponding results from our 2d models ( paper i ) after the nonlinear development of ti . at late times , however , these 3d turbulent models show much broader pressure distributions than our 2d models .
overall , the mean pressure averaged over orbits 6 - 10 is 1206 @xmath82 . by phase the mean
pressure is @xmath84 1187 , 1324 , and 1195 @xmath78 in the warm , intermediate and cold phases . also of interest
are the temperature pdfs , shown in figure [ temppdf ] at the same times as in figure [ denpdf ] . in panels c and d ,
the fraction of gas in the intermediate temperature phase has increased , and gas is also found at colder temperatures than are present earlier in the simulation .
the minimum temperature is 80k , and respectively 60% and 68% of the gas mass is found between 80 - 120k at @xmath77 and @xmath85 orbits . at @xmath62 and 9
orbits , on the other hand , respectively 30% and 18% of the gas is found at temperatures below 80k , while respectively another 32% and 31% of gas is at t=80 - 120 k. the range of temperatures in which the majority of cold gas is found increases by about a factor of two .
the upper limit on temperature increases slightly throughout the run , but in addition , the dispersion of temperatures in the warm medium increases . at early times
, @xmath8680% of the warm gas is in the range t=6600 - 8600k , whereas at late times , 80% is evenly distributed over twice as large a spread in temperatures .
figure [ tempcomp ] compares the volume - weighted temperature pdfs of four runs of different mean density .
these four runs have average densities of @xmath87 , and @xmath88 and , as we shall discuss in [ kinetics ] , the mean velocity dispersion increases by an order of magnitude from the highest to lowest mean density models .
the pdfs in figure [ tempcomp ] represent averages from 6.0 - 6.5 orbits . at intermediate and high temperatures , the pdfs for these runs
are quite similar .
most of the warm gas is at t=6000 - 8000 k , with @xmath89 10000 k. most of the cold phase is at temperatures near 100 k , possibly showing a slight trend towards higher mean temperature as @xmath90 is decreased .
overall there is less gas at lower temperature when @xmath90 is reduced , because the total mass available for cold clouds is lower .
in addition to having similar warm and cold gas temperatures , the models with various @xmath90 are similar in that the fractions of gas in the intermediate- and warm - temperature regimes are always quite close .
these results are illustrated in figure [ massfrac ] , which plots the mass fractions in the various regimes as a function of @xmath90 ( also including the @xmath91 model ) .
overplotted in figure [ massfrac ] are curves indicating the warm and cold gas mass fractions that a pure two - phase medium would have .
the mass fraction of cold gas in a perfect two - phase medium in thermal and pressure equilibrium is @xmath92 , where @xmath93 is the cold density , @xmath94 is the warm density , and @xmath8 is the mean density .
the mass fraction of warm gas is then @xmath95 .
the density of warm gas in our simulations is typically @xmath96 , which we use to compute the theoretical curves in figure [ massfrac ] .
the possibility exists that our choice of initial conditions in the standard run , a uniform medium at the average density , may have some effect on the amount of gas in the intermediate phase at late times .
due to ti , initially most of the gas collects into small , dense cold clouds , and only a small proportion of the gas remains in the thermally unstable regime .
later in the simulation , the mri drives a larger fraction of gas into the unstable phase .
it is possible that if we had _ begun _ with a turbulent medium , this fraction would be even larger , from increased shock heating of moderate density clouds with larger collision cross sections . to investigate this , we initialized a simulation with the same mean density and magnetic field as our standard run , but evolved it with an isothermal equation of state .
the sound speed was set so that the initial @xmath97 matches late time averages from our standard run .
after the isothermal evolution has proceeded for 10 orbits and reached a saturated turbulent state , heating and cooling are enabled . after a quasi - steady state
is reestablished , we measure the mass fractions in the warm , intermediate , and cold regimes .
the result is respective proportions of about 11% , 14% , and 75% , which is similar to our results from standard run .
thus , we conclude that the long - term thermal history does not strongly affect the present state of the gas .
in figure [ mach ] we plot the mass - weighted mach number @xmath98 of the gas in each thermal phase ( warm , intermediate , cold ) as a function of time for the duration of the simulation .
we also include , for comparison , the mass - weighted mach number of the cold medium for the high resolution run at @xmath33 .
the isothermal sound speed @xmath99 is computed individually for all grid zones , and the galactic shear is subtracted from the azimuthal ( @xmath100 ) velocity before computing @xmath101 .
initially , motions in all three phases of the gas are subsonic , @xmath102 , and remain so until the mri begins to develop at about 800 myr ( @xmath103 orbits ) .
once the mri saturates ( at @xmath56 orbits ) , the typical mach numbers of the warm , intermediate and cold phases of the gas are 0.4 , 1.8 , and 2.9 .
the peak value of @xmath104 for the cold phase is about 3.2 .
the mean late time velocity dispersion for all three phases of the gas is similar , approximately @xmath105 . at late times , the individual velocity dispersions in the radial , azimuthal , and vertical directions are 1.9 , 1.7 , and 0.7 @xmath106 , respectively . to explore the dependence of saturated state turbulence on system parameters , for our five simulations of varying mean density @xmath90 we have computed the average mach number over @xmath107 orbits .
we plot the results , separating the three thermal phases , as a function of @xmath90 in figure [ machplot ] .
the relationships between @xmath108 and @xmath8 clearly follow power laws .
the slopes for the warm , intermediate and cold phases are @xmath109 and -0.77 . since the cold component dominates the mass , this implies @xmath110 overall . for our @xmath111 model at @xmath112 , the saturated state mach numbers are 0.3 , 1.1 , and 1.6 for the warm , intermediate , and cold phases .
our results are thus consistent with general findings from previous mri simulations that saturated - state turbulent amplitudes increase with increasing mean alfvn speed .
the detailed scalings , however , show interesting differences , which we shall discuss in [ discuss ] .
we have found that the turbulence is quite insensitive to particularities of structure in initial conditions .
thus , our model which began with a two phase `` cloudy '' medium , with the same initial mean density as our standard run , saturates with nearly the same velocity amplitude as the standard run .
the initially - isothermal run which was restarted with cooling also yielded similar results to the standard run , with mach numbers of 0.4 , 1.7 , and 2.8 for the warm , intermediate and cold phases .
the saturated state of the isothermal simulation itself has a mach number of 1.4 , corresponding to mean velocity dispersion @xmath113 , somewhat larger than for our cooling models at this fiducial mean density .
differences between isothermal and multiphase models are likely to depend on @xmath90 , however .
the average reynolds stress , @xmath114 , from @xmath107 orbits is plotted against the mean density for @xmath115 and 0.25 @xmath116 in figure [ reynoldsplot ] .
the relationship again follows a power law , with a slope of -1.1 .
the velocity power spectra are generally consistent with previous simulations of the mri @xcite .
the largest scales dominate the simulation , generally following a kolmogorov - like spectrum , @xmath117 .
our quoted values for the velocity dispersions therefore correspond to the largest scales in the simulations . on smaller scales , such as an individual cloud
, the velocity dispersion would be smaller .
we have tested the relation between linewidth and size directly , using the `` roc '' analysis approach described in @xcite . both for the cold component alone , and for the whole medium
, we find that the velocity dispersion increases with the size of clouds , or sub - boxes of the computational volume . similarly to the ( random ) kinetic energy , the magnetic energy increases as the mri develops . in figure [ bfield ]
we plot the magnetic field strength as a function of time for each of the three phases of gas . in the initial conditions , @xmath118 g .
after ti develops , the field strength is @xmath119 g for the warm phase , and about @xmath120 g for the ( denser ) unstable and cold phases .
as the mri develops , after @xmath62 orbits , the field strength grows to range over @xmath121 g for all three phases , reaching as high as @xmath122 g in the cold phase .
the late time component magnetic field strengths , @xmath123 and @xmath124 are 1.3 , 1.9 , and 0.51 @xmath125 g , averaged over @xmath126 orbits .
thus , the mri enhances the magnetic field by an order of magnitude over its initial value .
we note that if overdense clouds were to form by isotropic contraction of the ambient medium , then one would expect @xmath127 . with a cold medium density two orders of magnitude larger than that of the warm medium , the respective mean field strengths would differ by a factor 20 . since this is not the case
, condensation evidently proceeds preferentially along field lines . to explore dependence on mean properties , in figure [ magplot ]
the late time magnetic field strength , averaged over five orbits , is plotted against the mean density in the box for five simulations with @xmath128 and 0.67 @xmath129 . unlike the turbulent velocity dispersions , the @xmath130 field strength does not show any significant trend with @xmath90 , saturating between 2 and 3 @xmath125 g .
the field strength also does not differ significantly between the cold , intermediate , and warm phases for any of the models . as a marginal effect , the field strength in the warm medium decreases as @xmath90 increases . unlike the magnetic energy density
, the maxwell stress does show dependence on @xmath90 .
this stress , @xmath131 , is averaged over @xmath107 orbits and plotted against the mean density for five simulations with @xmath132 and 0.25 @xmath116 in figure [ maxplot ] . for the data shown , a power law fit yields slope -0.42 .
previous single - phase mri simulations show somewhat different scalings of maxwell stresses and magnetic energies , as we shall discuss in [ discuss ] .
the power spectra of the magnetic field , like the velocity power spectra , is consistent with previous simulations of the mri @xcite , dominated by the largest scales and generally following a kolmogorov - like spectrum .
tracking the changes in various energies is key to understanding the interrelationships between dynamics and thermodynamics in turbulent flows .
for the models we have performed , the ultimate energy source is the shear flow , which drives the mri . in turn
, turbulent dissipation can convert kinetic and magnetic energy to thermal energy , which can subsequently be lost to radiation .
more formally , following hgb1 , we consider the average over the box of the total energy per unit volume , @xmath133 changes to this energy can occur due to losses or gains from radiation , and from fluxes through and stresses on the surface of the computational volume . with shearing - periodic boundary conditions ,
the net rate of change should ideally obey @xmath134 thus , if quasi - steady state is reached , we would then expect @xmath135 , and the sum of stresses times @xmath136 to equal the cooling rate . in steady state , from equation ( 3 ) the total rate of work done by the combination of compressions and shocks , @xmath137 , plus any other dissipation , should also be balanced by the net cooling , @xmath138 . in the upper panel of figure [ energy1 ] we plot ( for our another realization of our standard run ) the rate of work done by reynolds and
maxwell stresses per unit volume .
the late - time volume - averaged energy inputs from maxwell and reynolds stresses are 3.7 and 0.6 ( in units of @xmath139 ) .
the maxwell stress dominates the reynolds stress , which is typical in simulations of the mri . in the lower panel of figure [ energy1 ]
we plot the volume averaged shock heating , radiative heating - cooling rates , @xmath140 , and pressure work , @xmath141 , as a function of time .
the sum of these three terms is approximately zero during the first few orbits of the simulation . later in the simulation
there is typically either net heating or cooling at any particular time , but the late time averages - individually , 1.0 , -0.7 , and -0.3 ( in units of @xmath139 ) for shock , radiation , and pressure terms - sum to zero .
thus , on average , radiative losses exceed radiative gains , cooling by rarefactions exceeds heating by compressions , and together these net loss terms balance gains in shocks .
the mean energy density typically varies by 10% during the latter half of the simulation .
if total energy were perfectly conserved , as noted above , the energy inputs from maxwell and reynolds stresses would be balanced by net cooling .
the energy source for this radiative cooling would , in turn , be provided by compressive work and dissipation of turbulence .
the present simulation , however , in fact captures only part of the turbulent dissipation that in shock heating , as mediated by artificial viscosity .
in addition , both magnetic and kinetic turbulent energy are lost at the grid scale .
oppositely - directed magnetic fields and shear flows , when advected into a single zone , are averaged to zero .
since zeus evolves a ( non - conservative ) internal energy equation rather than a total energy equation , the associated energy from those small - scale sheared @xmath142 and @xmath143 fields is lost . in principle , these dissipation terms could be captured if explicit resistivity and shear viscosity were included . although the present simulations are highly idealized in many ways ( e.g. they are vertically periodic rather than stratified ) , it is interesting to explore model properties that bear a close relation to observables .
the profiles of 21 cm hi absorption directly trace the density , temperature , and turbulent velocities of the atomic ism via a line - of - sight convolution . using our simulated `` data
, '' we can generate analogous maps of line profiles projected in any direction through the computational volume .
figures [ line_x ] , [ line_y ] , and [ line_z ] show synthetic emission profile maps for our standard model along the x , y , and z directions .
we also present , paired with each line - of - sight velocity profile , the corresponding distribution of total emission with line - of - sight position .
each of the @xmath144 windows on the map represents a volume of @xmath145 zones , integrated over the projected area . for each zone ,
the contribution to emission is proportional to the density , with a gaussian velocity distribution centered on the flow velocity , and dispersion @xmath146 .
strong lines indicate more total mass along a given line of sight , and weak lines indicate less mass . since most of the mass in our standard
model is in the cold phase , a strong line indicates the presence of cold , dense gas . for figures [ line_x ] , [ line_y ] , and [ line_z ]
the mean line widths of the velocity profiles are 1.8 , 1.9 , and 1.7 @xmath147 . without thermal broadening the line widths
are reduced to 1.0 , 1.3 , and 0.7 @xmath148 , which is consistent with the time - averaged velocity dispersions in the radial , azimuthal , and vertical directions reported in [ kinetics ] .
most of the velocity profiles are single - peaked , and would likely be interpreted as arising from one to three emitting components if a standard gaussian fitting procedure were applied .
however , our results show that in many cases several spatially - separated components are distinctly evident in the line - of - sight mass distributions .
velocity profile broadening and skewness statistically give evidence that more than one component is present , but we find no correlation between increased spatial coherence and increased velocity profile symmetry in any given direction .
we do not observe any structure which shows two distinct lines .
the absence of velocity profiles with two distinct peaks owes in part to the thermal broadening , which smears out smaller scale features .
however , the primary reason that profiles are single - peaked is that velocity modes at a large range of wavenumbers are present in saturated state mri - driven turbulence .
since there is no single dominant wavelength along the line - of - sight , the range of velocities is smoothly filled .
the spatial resolution of the synthetic observations was doubled to determine if the structure of the velocity profiles would be affected .
generally , the line profiles remain single - peaked , whether or not well - separated spatial components are present . in figure [ line_x_warm ]
we show velocity profiles for the warm gas only , and do not include thermal broadening . the line - of - sight position profile ( also in figure [ line_x_warm ] )
shows that the warm gas is spatially much more uniformly distributed than the cold gas , which dominates the profiles in figures [ line_x ] - [ line_z ] .
interestingly , however , the intermediate - temperature gas is always associated with cold condensations .
this is clearly seen in figure [ phaseplot ] , which shows slices through the volume both before and after the onset of strong turbulence .
the line profiles for runs with different mean density are very similar to the standard run .
they typically show a single component with occasional evidence for a weaker second component . as the mean density
is decreased to @xmath149 and @xmath150 , the line widths increase to 2.0 , 2.2 , and 1.9 @xmath148 , and 4.8 , 4.7 , and 4.5 @xmath148 , respectively . without thermal broadening the line widths
are reduced to 1.1 , 1.5 , and 0.74 @xmath12 , and 2.2 , 2.2 , and 1.1 @xmath12 , respectively . a similar trend of decreasing line width with increasing mean density is also observed .
in this paper , we present results from a set of numerical mhd simulations that focus on the interrelationship between turbulence and thermal structure .
the models we have performed are three dimensional , and include sheared galactic rotation and magnetic fields .
turbulence therefore is generated by the magnetorotational instability .
we also include a radiative cooling function that , in pressure equilibrium , would yield a two - phase medium .
the two fundamental issues we have addressed are ( 1 ) how cloudy structure alters the saturated - state properties and scalings of mri - driven turbulence , compared to single - phase mri models , and ( 2 ) how turbulence that is _ not _ driven by direct ( stellar ) thermal energy inputs affects the thermal balance and phase structure in the warm / cold atomic medium .
_ evolution and physical structure : _ a two - phase cloudy medium with many small clouds develops in the first 20 myr of our simulations . over time , due initially to galactic shear , and later ( @xmath151 orbits ) to mri - driven turbulence , these clouds undergo a continual series of mergers and disruptions , leading to a late - time state in which the mass function of condensations peaks at a few hundred @xmath152 .
the dense condensations are triaxial , and typically have max : min axis ratios of 2:1 .
they consist of cold gas lumps surrounded by envelopes of thermally - unstable gas ; filling all of the remaining volume is thermally - stable warm gas .
_ density and temperature distributions : _ for the range of parameters we have explored , in late stages of evolution most of the gas mass is in the cold phase , while most of the volume is occupied by the warm phase . while the proportion of thermally - unstable intermediate - temperature gas in a given model increases after the advent of mri , at all stages the density and temperature pdfs show distinct warm and cold phases , with a varying amount of material in the `` non - equilibrium '' valley between these peaks .
the peaks , near @xmath153 and @xmath154 , also broaden as the turbulence develops .
the relative proportions in each phase depend on the mean density , varying from 95% cold gas when the mean density @xmath155 , to 50% cold gas when @xmath156 .
the fractions of thermally - unstable gas and warm gas are always comparable to each other . increasing levels of turbulence yield increasing proportions of thermally - unstable gas .
relative to the proportions predicted for a two - phase , quiescent medium in thermal and pressure equilibrium , increasing turbulence also tends to _ increase _ the fraction of cold gas , while decreasing the fraction of warm stable gas .
_ pressure : _ we initialize our models at @xmath157 , but secular cooling in the stages before mri develops leaves the gas in approximate pressure equilibrium ( @xmath158 ) at a lower mean pressure of @xmath159 , near the minimum for which two stable gas phases can be present .
after mri develops , pressures cover a much wider range of values ( @xmath160 ) , with a maximum at @xmath161 , but relatively unchanged mean value ( @xmath162 ) .
_ turbulent velocities : _ after the mri saturates at @xmath163 orbits , the turbulent velocity dispersion reaches a quasi - steady plateau albeit with fluctuations of @xmath164 in amplitude . for our fiducial model with @xmath165 , the mean late time ( 3d ) velocity dispersion is @xmath166 for all three components .
this velocity corresponds to mean mach numbers of 0.4 , 1.8 and 2.9 in the warm , intermediate , and cold phases .
we examined the effect of mean density on the velocity dispersion , and found that @xmath167 overall , with slightly shallower slope for the warm gas alone .
our results show , additionally , that the reynolds stress , @xmath168 , varies with mean density @xmath169 .
we find that the in - plane components of the velocity dispersion exceed the component perpendicular to the disk by about a factor of two .
_ magnetic fields : _ for the present set of models , we have adopted initial conditions with a uniform vertical magnetic field of strength @xmath170 g .
the mri enhances the field by an order of magnitude , so that @xmath171 is typically @xmath121 g late in the simulation .
the field strength is similar ( within @xmath172 ) in all three phases of gas , and there is no significant trend of field strength with mean simulation density @xmath8 .
however , we find that the maxwell stress , @xmath173 , varies with mean density @xmath174 .
synthetic line profiles : _ as a demonstration of the potential for employing simulations to interpret observational diagnostics , we have computed maps of synthetic line profiles from sample data cubes . we find that the line profiles are generally single - peaked ( although in some cases would require two or three components if a standard gaussian fitting scheme were performed ) . in no case did we identify two distinct velocity components , even though there are distinct cloud structures present along many lines of sight . because turbulence has a smooth power spectrum ,
this kind of overlap in velocity space is inevitable .
our results have several interesting implications for interpreting ism observations , and it is also interesting to compare with recent numerical and theoretical work on the ism and on mri dynamics .
we conclude by discussing these connections .
high levels of turbulence are observed both in the atomic gas of the milky way , and in that of external spiral galaxies , and it has been suggested that the mri could be an important contributor to this turbulence , especially in the outer parts of galaxies where there is little star formation .
our models are the first ( to our knowledge ) to address this issue directly with an appropriate physical model
namely , one that admits two stable thermal phases , such that the mri must develop in a cloudy medium with density contrasts of 100 between clumps and diffuse gas . while the turbulent velocity that develops in our fiducial model with mean density @xmath165 is relatively modest , the scaling of the turbulent amplitude with @xmath175 is quite steep , such that @xmath176 is predicted when @xmath177 .
the scaling of turbulent velocity dispersion with mean density indicates that mri may play a significant role in the outer regions of the galaxy . beyond the point in the milky way where the stellar surface density drops , the gas scale height rapidly increases , and the volume density correspondingly decreases ; this sort of disk flaring is also seen in external galaxies . in the @xcite milky way model , for example
, @xmath8 falls below @xmath178 at @xmath179 .
the outer - galaxy pressure in the @xcite model is nevertheless high enough for cold - phase gas to be present , so that if it were _ not _ turbulent , then a thin gravitationally unstable layer would develop . ( the threshold for instability ) for @xmath180 , where @xmath181 is the gas surface density .
@xcite show that cold gas is expected to be present at all radii out to 18 kpc . since @xmath181 is estimated to exceed @xmath182 inside this radius , even if only 20% of the gas is in the cold component , the layer would be gravitationally unstable . ]
our results suggest that the mri could be maintaining high - amplitude turbulence , and hence suppressing star formation , in the far outer milky way and other spiral galaxies .
we note that the increase in scale height is necessary for this to hold ; since the minimum mri wavelength @xmath183 , mri - driven turbulence can only be sustained in a sufficiently thick disk . even in the inner disk ,
our results suggest that mri may be a significant contributor to turbulence in the ism . at a mean inner - galaxy ( @xmath184 ) midplane density of @xmath185 @xcite
, our results would predict @xmath186 . away from the midplane where the density drops
, the turbulent amplitudes would increase .
the mean inner - disk vertical magnetic field strength may also be somewhat larger than the fiducial value we have adopted ( @xcite obtained @xmath187 g from pulsar observations ) , which would tend to increase the amplitude of the turbulence .
a more extensive parameter survey allowing for disk stratification , varying scale height , and differing initial field strengths and distributions is needed to quantify more fully the expected contribution from mri to turbulent amplitudes in the ism .
another important question is whether mri development could be quantitatively altered by interaction with large - scale pertubations driven by supernovae or spiral shocks .
we defer consideration of this interesting issue to future work
. direct comparisons between simulations and observations regarding levels of turbulence and magnetic field strength as a function of local parameters would be very useful .
unfortunately , observations at this time do not permit such comparisons to be made .
hi velocity dispersions in the milky way can only be measured within @xmath86 1 kpc of the sun ( e.g. @xcite ) . in external
near - face - on galaxies , observed ( vertical ) velocity dispersions combine both turbulent and thermal contributions , and these values do not vary secularly with galactic radii ( although dispersions are significant ) even well beyond the optical disk ( see 1 ) . magnetic field strengths in the milky way beyond @xmath86 10 kpc
have not been measured directly ( i.e. with faraday rotation ; see e.g. @xcite ) .
for both the milky way and external galaxies , one may use synchrotron emission @xcite to obtain the product of the magnetic and cosmic ray energy densities as a function of galactic radius , but since the equipartition assumption need not be satisfied everywhere , this does not yield a b - field strength except locally , where electron cosmic - ray and gamma - ray observations can be made . milky way
outer - galaxy field strengths of 2 - 3 @xmath125 g are consistent with synchrotron / cosmic ray models of @xcite .
the scalings we find for mri amplitudes show interesting differences from those obtained with single - phase gaseous media , in previous adiabatic and isothermal simulations . in the shearing - box models of hgb1 , hgb2 , and @xcite , all of the measures of turbulence scale together i.e. @xmath188 . for simulations with net vertical magnetic flux ,
hgb1 further reported that these stresses and energy densities scale @xmath189 , where @xmath190 and @xmath47 are the initial alfvn speed and sound speed , respectively .
@xcite , on the other hand , identify a scaling of stresses @xmath191 in the mean alfvn speed , and also demonstrate that the pressure dependence of the saturated - state stresses are very weak . while we have not surveyed cases with differing magnetic field strengths , our models at varying mean density have varying @xmath190 . at fixed mean @xmath192
, the single - phase medium simulations cited above would predict scalings @xmath193 or @xmath194 for @xmath7 , @xmath195 , and the stresses . for our cloudy - medium simulations ,
we in fact find that @xmath7 is nearly independent of @xmath175 , while other scalings in our models are either in the same range as the single - phase predictions ( i.e. @xmath196 and @xmath197 ) , or slightly steeper ( @xmath198 ) .
interestingly , 3d mri simulations in radiation - dominated disks @xcite , which like our models contain strong density contrasts , also show a steep dependence of the stress @xmath199 .
the reason for the difference between single - phase and cloudy - medium results for saturated - state @xmath7 scalings is not yet clear , but the lack of dependence on @xmath8 in our models suggests that local , rather than global , properties of the gas determine the field strength that develops .
the densities in the diffuse phase and in the dense phase are similar for all our models ; only the filling factor of cold clouds differs appreciably .
if the saturated - state field strength depends on the reconnection rate , and this depends on local gas densities and field geometry , then the fact that these properties are the same in all our two - phase models might explain the lack of dependence of @xmath7 on @xmath8 . our isothermal model , which has less - kinked magnetic field , and typical ( log ) densities midway between @xmath200 and @xmath201 , indeed has saturated - state values of the rms field strength 20 - 50% higher than the standard run ( @xmath202 g in the isothermal model , compared to 2.4 , 2.7 , and 2.9 @xmath125 g in the f , g , and h phases for the standard run ) .
further study , with particular focus on the rate and spatial distribution of reconnection , is needed to clarify this issue . while the saturated - state magnetic field strength depends on a balance between mri - driven amplification and ( numerical or true resistive ) dissipation , the saturated - state velocity dispersion depends on a balance between mri driving and losses in shocks , compressions , and ( numerical or true viscous ) shear dissipation . since turbulent velocities are similar in all the gas components , while the mass is concentrated in the cold clouds , in paper i we proposed that the cloud collision time may be a good proxy for the kinetic energy dissipation time .
the kinetic energy dissipated per unit time per unit mass is then @xmath203 , where @xmath204 is a cloud size and @xmath205 is the gas density within clouds .
the kinetic energy input rate per unit mass due to mri is an order - unity constant times @xmath206 .
balancing inputs with dissipation , and using @xmath207 from our simulations , this predicts a scaling for the saturated - state velocity dispersion @xmath208 .
this prediction indeed agrees well with the velocity dispersion scaling measured directly from our simulations , @xmath209 .
of course , this scaling can not continue to arbitrarily low density , because the mri becomes stabilized if the wavelength @xmath210 exceeds the height of the disk .
in previous analytic work on magnetized cloud - cloud collisions @xcite argued that effective cloud cross sections should vary @xmath211 .
although we do not measure this effect directly , it would not be expected to change our results significantly . taking this into account for @xmath204 in the above analysis , the predicted power - law scaling exponent for @xmath5 with @xmath212 changes by only 0.02 . as noted above
, our results show only minor differences between typical magnetic field strengths in gas of different phases , with the cold medium having slightly higher @xmath130 .
if flux were frozen to the gas , and cold clouds contracted isotropically out of the warm ism due to ti , then the field strengths would differ by a factor 20 between diffuse and dense phases . evidently , however , this is not the case : observationally , @xcite reports the @xmath130-fields in cold atomic clouds are no stronger than in the diffuse ism overall .
our models are consistent with this result , in part because clouds in fact do not condense isotropically ( but instead preferentially gather material along field lines ) , and in part because field lines can diffuse relative to matter ( in our models , this is purely numerical , but turbulent diffusion is likely to play the same physical role ) .
synthetic line profiles of the sort we have computed are potentially of great value in interpreting observations of the ism , and in particular , the emission and absorption profiles of the 21 cm hydrogen line .
ensemble properties of the atomic ism , including the turbulent velocity dispersion for the cold gas separately , can be unambiguously obtained from observed position - velocity data cubes .
our results illustrate the difficulty , however , in discerning detailed ( local ) structural properties of the ism directly from line profiles , due to overlapping in velocity space ( cf .
ostriker , stone , & gammie 2001 ) .
fortunately , recent work has demonstrated that analysis techniques calibrated using simulations can be used to extract fundamental statistical properties such as power spectra from co observations of turbulent molecular clouds ( e.g. @xcite ) , and it will be interesting to test whether the same holds true for hi gas .
what do we conclude regarding the dynamics / thermodynamics connection in the turbulent atomic ism ? as alluded to in 1 , recent observations @xcite have suggested that at least 30% of the hi gas at high latitudes ( where blending due to galactic rotation is minimized ) is in the thermally - unstable regime ( @xmath213 k ) .
another 30% of the hi mass at high latitudes is estimated to be warm , thermally - stable gas.k in the @xcite data , suggesting that midplane gas may be much more likely to be in thermal equilibrium . ]
interestingly , we find that in our simulations , the warm and thermally - unstable gas mass fractions are also always very similar . in our models ,
the warm+unstable mass fraction only approaches 50% for the lowest-@xmath175 case ; at @xmath214 , the warm+unstable gas comprises 30% of the total mass .
however , except at the lowest @xmath175 , the turbulent amplitudes in our models are also somewhat lower than in the real ism , so in part the lower warm+unstable fractions we find may owe to lower rates of turbulent dissipation . in addition , as discussed in 3.6 , because our numerical method is non - conservative , a significant fraction of the input turbulent energy from the mri is lost without being thermalized .
our method therefore underestimates the heating rate
. this may be especially true in high density regions where the magnetic field curvature is large , so that the current density @xmath215 is large , and the resistive heating @xmath216 should also be large . we intend to explore this issue in future work incorporating explicit resistivity , also comparing with results using a conservative algorithm to update the energy equation .
overall , our results are consistent with the recent work of @xcite , who performed high resolution 2d simulations of a converging flow in which turbulence develops , using a cooling function very similar to ours .
the temperature and density pdfs from their models are qualitatively similar to ours , in particular showing evidence for the existence of a two - phase medium even in the most turbulent case .
similar to our results , they found that the proportion of thermally - unstable gas increases with the amplitude of turbulence .
for their least turbulent model , about 10% of the gas was thermally unstable , while this fraction increased to 30% in models with larger turbulent amplitudes .
other recent work , based on simulations that include modeled effects of star formation via prescriptions for ob star heating ( e.g. @xcite ) and supernovae ( e.g. @xcite ) , have found significant fractions of gas in the unstable phase , and many have advocated for a shift away from the classical two- or three - phase medium concept towards more of a `` phase continuum '' description
. however , the former set of models have energy inputs from `` star formation '' in cold atomic gas ( rather than in self - gravitating molecular clouds ) , which likely leads to overestimating the amount of thermally - unstable gas .
the latter models have not included low - temperature cooling , so no cold phase can form .
thus , it is not yet clear whether realistic models incorporating turbulence driven by star formation would indeed produce an extended , featureless continuum of temperatures , or whether they would yield bimodal distributions similar to those we have found with solely mri - driven turbulence .
we believe that , given the observed turbulent amplitudes ( @xmath217 ) and characteristic spatial scales ( the disk thickness @xmath218 pc ) for the atomic ism , quasi - two - phase distributions are inevitable .
the heating time @xmath219 yr is a factor of a few longer than the cooling time in gas at densities @xmath220 ( see eq . [ 5 ] of paper i or eq .
[ 4 ] of @xcite ) , so that moderate but not extreme departures from thermal equilibrium can be expected in diffuse gas . in dense gas ,
cooling times are very short , so thermal equilibrium must hold , but typical order - unity variations in pressure from sonic - level turbulence can lead to moderate local density and temperature variations .
future work will be able to determine whether , with realistic turbulent amplitudes and fully - captured energy dissipation , thermal distributions from models can indeed match those from hi observations , or whether additional heat sources are required .
we are grateful to charles gammie , woong - tae kim , jim stone , and mark wolfire for valuable discussions , and yen - ting lin and jonathan mckinney for helpful technical advice .
the manuscript has also benefited from a thoughtful and detailed report prepared by an anonymous referee .
this work was supported in part by grants nag 59167 ( nasa ) and ast 0205972 ( nsf ) .
some of the computations were performed on the tungsten cluster at ncsa , and others were performed on the ctc cluster in the umd department of astronomy . | this paper reports on three - dimensional numerical simulations of dynamics and thermodynamics in the diffuse interstellar medium ( ism ) .
our models are local , account for sheared galactic rotation , magnetic fields , and realistic cooling , and resolve scales @xmath0 pc .
this combination permits the study of quasi - steady - state turbulence in a cloudy medium representing the warm / cold atomic ism .
turbulence is driven by the magnetorotational instability ( mri ) ; our models are the first to study the saturated state of mri under strongly inhomogeneous conditions , with cloud / intercloud density and temperature contrasts @xmath1 . for volume - averaged densities @xmath2 , the mean saturated - state velocity dispersion ranges from @xmath3 , with a scaling @xmath4 .
the mri is therefore likely quite important in driving turbulence in low - density regions of the ism , both away from the midplane in the inner galaxy ( as observed at high latitudes ) , and throughout the far outer galaxy ( where the mean density drops and the disk flares ) .
the mri may even be key to suppressing star formation at large radii in spiral galaxies , where the pressure can be high enough that without mri - driven turbulence , a gravitationally - unstable cold layer would form .
as expected , we find that turbulence affects the thermal structure of the ism . in all our simulations , the fraction of thermally - unstable gas increases as the mri develops , and in the saturated state is largest in high-@xmath5 models .
the mass fractions of warm - stable and unstable gas are typically comparable , in agreement with observations .
while inclusion of resistive dissipation of magnetic fields could enhance the amount of thermally - unstable gas compared to current models , our present results indicate that even high levels of turbulence can not wipe out the signature of thermal instability , and that a shift to a `` phase continuum '' description is probably unwarranted . instead
, we find that temperature and density pdfs are broadened ( and include extreme departures from equilibrium ) , but retain the bimodal character of the classical two - phase description .
our presentation also includes results on the distribution of clump masses ( the mass spectrum peaks at @xmath6 ) , comparisons of saturated - state mri scalings with single - phase simulation results ( we find @xmath7 is independent of @xmath8 ) , and examples of synthetic hi line profile maps ( showing that physical clumps are not easily distinguished in velocity components , and vice versa ) . |
with rising demand for bandwidth , several researchers around the world have measured and studied the occupancy of spectrum in different countries .
these measurements suggest that except for the spectrum allocated to services like cellular technologies , and the industrial , scientific and medical ( ism ) bands , most of the allocated spectrum is heavily underutilized .
the overall usage of the analyzed spectrum is as low as 4.54% in singapore @xcite , 6.2% in auckland @xcite , 17.4% in chicago @xcite and 22.57% in barcelona @xcite . among all the unutilized portions of the frequency spectrum , white spaces in the ultra high frequency ( uhf ) television ( tv ) bands have been of particular interest owing to the superior propagation characteristics as compared to the higher frequency bands .
loosely speaking , the unutilized ( or underutilized ) tv channels collectively form the tv white spaces .
the amount of available tv white space varies with location and time .
tv white space estimation has been done in countries like the united states ( us ) , the united kingdom ( uk ) , europe , and japan @xcite . in the indian context
, single - day experiments at three locations in urban and sub - urban delhi have been performed @xcite .
the estimation of tv white space in the uhf band , based on spectrum allocation and tv transmitter parameters , is presented in this work .
the main contributions of this paper are the following : 1 .
for the first time , the empirical quantification of the available tv white space in the @xmath0-@xmath1mhz in india is presented .
the quantification utilizes existing methods in the literature , namely pollution and protection viewpoints @xcite , and the technical specifications of the federal communications commission @xcite .
it is found that uhf tv band spectrum is heavily underutilized in india .
2 . motivated by underutilization of uhf tv band spectrum
, a spatial reuse based channel allocation algorithm has been proposed for the existing indian tv transmitters operating in the 470 - 590 mhz band .
the algorithm uses the least number of tv channels while ensuring no ( significant ) interference between transmitters operating in the same channel .
it is observed that at least @xmath4 uhf tv band channels can be freed by this approach .
the importance of the above results must be understood in the context of indian national frequency allocation plan ( nfap ) 2011 where a policy intent for the utilization of tv white spaces was made .
therefore , it is necessary to estimate the amount of tv white spaces in india . besides , based on above results ,
the tv band in india is underutilized and this situation is quite different than in the developed countries .
the optimal mechanism(s ) for the use of tv white spaces in india can be _ different _ and it should be studied by further research . _
organization : _ the tv white space scenario and the related work on quantitative analysis in a few other countries is briefly described in sec .
[ sec : tvws_review ] .
[ sec : india_tvplan ] describes the current indian usage scenario of the uhf tv bands .
[ sec : methodology ] presents the methodology and assumptions used in calculating the white space availability in india .
[ sec : results ] presents the results of our work , and compares the tv white space availability in india with that of other countries . in sec .
[ sec : channel_allocation ] , we propose a frequency allocation scheme to the tv transmitters in india so as to ensure minimum number of channel usage in the country .
concluding remarks and directions for future work are discussed in sec .
[ sec : conclusions ] .
regulators fcc in the us and ofcom in the uk have allowed for secondary operations in the tv white spaces . under this provision , a secondary user can use the unutilized tv spectrum provided it does not cause harmful interference to the tv band users and it relinquishes the spectrum when a primary user ( such as tv transmitter ) starts operation . since the actual availability of tv white spaces varies both with location and time , operators of secondary services are interested in the amount of available white space .
the available tv white space depends on regulations such as the protection margin to the primary user , maximum height above average terrain ( haat ) , transmission power of secondary user , and the separation distance . as per fcc
, a band can declared as unutilized if no primary signal is detected above a threshold of @xmath5dbm @xcite . using the parameters of terrestrial tv towers ,
tv white space availability in the us has been done in the literature @xcite .
the average number of channels available per user has been calculated using the pollution and protection viewpoints .
these viewpoints are explained in more detail in sec .
[ sec : methodology ] . using the pollution viewpoint into account , the average number of channels available per location increases with the allowable pollution level .
this average number of available channels is maximum in the lower uhf band . in the protection viewpoint too , the average number of available channels at a location is maximum in the lower uhf band ( channels 14 - 51 of the us ) and this decreases as more and more constraints are applied . in uk , ofcom published a consultation providing details of cognitive access to tv white spaces in 2009 @xcite .
the coverage maps and database of digital tv ( dtv ) transmitters can be used to develop a method for identification of the tv white space at any location within uk @xcite . the tv white space availability in japan has also been studied in @xcite .
the results of @xcite indicate that the amount of available tv white space in japan is larger than that in us and uk .
however , this availability decreases with an increase in the separation distance . to the best of our knowledge ,
a comprehensive study of tv white space availability has not been done in india and is the focus of this work .
as per the nfap 2011 @xcite , the spectrum in the frequency band @xmath0-@xmath6mhz is earmarked for fixed , mobile and broadcasting services .
the nfap has allowed the digital broadcasting services to operate in the @xmath7-@xmath8mhz band .
india is a part of the itu region 3 , and the @xmath8-@xmath9mhz band has been earmarked for international mobile telecommunications - advanced ( imt - a ) applications ( see footnote ind 38 of @xcite ) .
hence , the digital tv broadcasting will operate in the frequency band from @xmath7mhz to @xmath8mhz .
currently the tv transmitters operate only in the @xmath0-@xmath1mhz band in the uhf band . in india , the sole terrestrial tv service provider is doordarshan which currently transmits in two channels in most parts across india .
currently doordarshan has @xmath10 tv transmitters operating in india , out of which @xmath11 transmitters transmit in the vhf band - i ( @xmath12-@xmath13mhz comprising of two channels of @xmath14mhz each ) , @xmath15 transmitters operate in the vhf band - iii ( @xmath16-@xmath17mhz comprising of eight channels of @xmath11mhz each ) , and the remaining @xmath18 transmitters transmit in the uhf band - iv ( @xmath0-@xmath1mhz comprising of fifteen channels of @xmath11mhz each ) . in india ,
a small number of transmitters operate in the uhf bands . as a result ,
apart from @xmath11-@xmath19mhz band depending on the location , _ the uhf band is quite sparsely utilized in india _ !
this observation will be made more precise in the next section .
the quantification of tv white space in india will be addressed in this section .
a computational tool has been developed that calculates the protection region and separation distance for each tower , and also the pollution region around the tower where a secondary device should not operate .
currently , there are no tv white space regulations in india . the regulations of fcc ( us ) are borrowed for the estimation of tv white space in india .
_ microphones are ignored in our computation due to lack of available information_. the input to the developed computational tool include the following parameters for all the tv transmitters : 1 .
position of the tower ( latitude and longitude ) , 2 .
transmission power of the tv transmitter , 3 .
frequency of operation , 4 .
height of the antenna , 5 . and , terrain information of area surrounding the tower .
the above parameters of all the tv towers operating in the uhf band - iv have been obtained from the terrestrial tv broadcaster doordarshan . out of 373 transmitters operating in the 470 - 590 mhz uhf band ( channel no .
21 - 35 ) , doordarshan has provided data of 254 towers operating in the west , east , south and the north east zone .
the tv transmitter information for the north zone is yet to be provided by doordarshan .
comprehensive field strength measurements in india suggest that hata model is fairly accurate for propagation modeling @xcite .
the hata model will be used for path - loss calculations . using the tv transmitter information and the propagation model
, we quantify the available tv white space in the uhf tv band by two methods .
the first method utilizes the protection and pollution viewpoints while the second one utilizes technical specification made by the fcc . the protection and pollution viewpoints used for calculating tv white space
have been introduced by mishra and sahai @xcite .
their method is reviewed in this section and utilized in our work for obtaining tv white space availability ( see sec . [
sec : results ] ) . in the protection viewpoint ,
when a secondary user operates , it must not cause any interference to the primary receivers in its vicinity .
this is illustrated in fig .
[ fig : picture31 ] protection radius , separation distance and the no - talk radius , width=240 ] the protected area is defined using the following @xmath20 equations .
let @xmath21 be the transmit power of primary in dbm , @xmath22 be the path - loss in db at a radial distance @xmath23 from the transmitter , @xmath24 be the thermal noise in dbm , and @xmath25 be the threshold @xmath20 in dbm .
then , the protection radius @xmath26 is defined by the following @xmath20 equation , @xmath27 the regulator provides an additional margin ( @xmath28 ) to account for fading .
the modified equation for @xmath26 is , @xmath29 the no - talk radius @xmath30 is defined as the distance from the transmitter up to which no secondary user can transmit .
the difference @xmath31 is calculated such that if a secondary device transmits at a distance of @xmath31 from the tv band receiver located at @xmath32 , the @xmath20 at the tv band receiver within a radius @xmath33 does not fall below @xmath25 .
the separation distance @xmath31 is then calculated such that @xmath34 where , @xmath35 is the secondary transmitter power in dbm .
in addition to the co - channel considerations , a tv receiver tuned to a particular channel has a tolerance limit on the interference level in the adjacent bands . in the protection viewpoint , we consider that the protection radius in the adjacent channel is the same as in co - channel
. however , the tv receiver can tolerate more adjacent channel interference than co - channel interference .
therefore , a margin of @xmath36db more than co - channel fading margin @xmath28 ( set by the fcc regulations @xcite ) is provisioned for adjacent channel interference .
the pollution viewpoint takes into consideration the fact that even though a region could be used by a secondary device , the interference at the secondary receiver due to the primary transmitter might be higher than the tolerable interference level of the secondary receiver .
if @xmath37 is the interference tolerable by the secondary receiver , then @xmath38 is given by , @xmath39 similar to the protection viewpoint , there are adjacent channel conditions ( leakage of primary transmitter s power in the adjacent channel ) in the pollution viewpoint as well .
it is assumed that the secondary device can tolerate up to @xmath40db of interference if it is operating in the adjacent channels .
the tv white space available is the _ intersection _ of the white space determined from the pollution and protection viewpoints .
the parameters used in our computations for calculating the available tv white space are given in table [ i ] .
& @xmath41db + & @xmath42db ( specified for 802.11 g systems ) + maximum tolerable interference ( @xmath37 ) by secondary ( adjacent channel ) & @xmath40db + noise in a @xmath11mhz band ( @xmath43 ) & @xmath44dbm + + & @xmath45db + & @xmath46db ( specified by fcc ) + additional fading margin in adjacent channel & @xmath36db ( specified by fcc ) + required @xmath20 for primary receiver & @xmath40db + transmission power of secondary device & @xmath47dbm + haat of secondary device & @xmath48 m + as an example , we consider the tv tower located at the sinhagad fort in pune .
doordarshan informed us that the tower at sinhagad fort operates in the @xmath49-@xmath50mhz band ( channel @xmath51 ) at a height of @xmath2 m and power of @xmath52kw ( @xmath53dbm ) . in the hata model used for path loss calculations ,
pune has been considered as an urban city . using the pollution viewpoint , for a @xmath42db tolerable interference in channel @xmath51 ( @xmath49-@xmath50mhz )
, the pollution radius for the tower is calculated to be @xmath54 km , and for a tolerable interference of @xmath40db in the adjacent channel , the pollution radius is @xmath55 km .
what this means for a secondary device is that the interference level is more than the allowable limit ( @xmath42db above noise floor ) in a region of @xmath54 km in channel @xmath51 and @xmath55 km in the adjacent channels around the tower . from the protection viewpoint , if a fading margin of @xmath46db is provided , the protection and no - talk radius in channel @xmath51 are @xmath56 km and @xmath57 km respectively .
if we consider an additional fading margin of @xmath36db in the adjacent band , the no talk radius in the adjacent channel is @xmath56 km .
this implies that if a secondary device operates within a distance of @xmath57 km in channel @xmath51 and @xmath56 km in the adjacent channels , the primary user receiving on channel @xmath51 will experience interference .
the available white space is the intersection of the white space using the two viewpoints .
thus , in pune , no secondary device can operate within a distance of @xmath54 km ( limit set by pollution viewpoint ) on channel @xmath51 and @xmath56 km in the adjacent channels ( limit set by protection viewpoint ) around the tower at sinhagad fort . in the fcc s definition of tv white space , the protection radius is same in the grade b contour ( @xmath58 ) @xcite . in the uhf band , @xmath58 is the distance from the tv tower where the field strength of the primary signal falls to @xmath59dbu .
the required field strength is converted from from dbu to dbm using the following conversion formula @xcite , @xmath60 where , @xmath61 is transmit power in dbm , @xmath62 is the field strength in dbu , @xmath63 is the upper frequency - limit of the channel , and @xmath64 is the lower frequency - limit of the channel . to calculate the separation distance , i.e. distance beyond @xmath58 where no secondary device can transmit , the distance @xmath65 such that the signal from the secondary device at @xmath30 results in a signal level of @xmath66dbu at the tv receiver located at @xmath58 is calculated . for the tv transmitter at pune ,
the no - talk radius , i.e. the distance from the tower beyond which a secondary device can use the channel is computed to be @xmath67 km .
the results obtained by tv white space calculation methods of sec . [ sec : methodology ] will be discussed in this section . using the methodology described in sec .
iii , the pollution and the no - talk radius are calculated for every tv tower in the four zones .
each region is plotted as a circle around the tv tower .
here it has been assumed that each tower has an omnidirectional antenna . the tv white space availability in the west , east , south and north - east zone using the pollution viewpoint
is shown in fig .
[ poll15db ] and using the protection viewpoint in fig .
[ prot1db ] .
white space availability using the fcc regulations is shown in fig .
[ fcc2 ] .
[ poll15db ] and fig .
[ prot1db ] illustrate that at most places in india , not even a single channel in the uhf band is utilized ! to quantify this result further , the complementary cumulative distribution function of the number of channels available per unit area as tv white space is plotted in fig .
[ ccdf1 ] . [ cols="^,<,<,<,<,<,<",options="header " , ]
conclusions that can be drawn from table [ ii ] are as follows : 1 . out of the 15 uhf tv channels ( 470 - 590 mhz )
, the average number of tv channels available for secondary usage is above 14 ( 112mhz ) in each of the four zones .
2 . available
tv white space is the maxium in the north east , where 18 transmitters operate in the uhf band .
3 . if we use the adjacent channel constraint , the available white space decreases . however , this decrease is less than 1% in each case .
table [ iii ] concludes that in almost all cases at least 12 out of the 15 channels ( 80% ) are available as tv white space in 100% of the areas in india .
this is larger than japan @xcite , where out of 40 channels , on an average 16.67 channels ( 41.67% ) are available in 84.3% of the areas .
this white space is also larger than what is available in us and the european countries .
the available tv white space by area in germany , uk , switzerland , denmark on an average are 19.2 ( 48% ) , 23.1 ( 58% ) , 25.3 ( 63% ) and 24.4 ( 61% ) channels out of the 40 channels respectively @xcite . similarly , as compared to the us , the available tv white space in india is much larger .
it must be noted that in tv white space studies across the world , the imt - a band is also included .
there are a total of 254 doordarshan tv transmitters in the four zones illustrated in fig .
[ loc ] operating in the @xmath0-@xmath1mhz .
currently , in these zones , 14 out of the 15 channels ( channels @xmath68-@xmath69 ) are _ sparsely _ used for transmissions . as shown in fig .
[ loc ] , channels allocated to the transmitters are reused inefficiently or at very large distances . for instance , out of the @xmath70 transmitters , only @xmath71 transmitters in the four zones operate on channel @xmath68 .
we propose a channel allocation scheme such that the minimum number of tv channels are used in each zone , while ensuring that the coverage areas of different transmitters do not overlap .
the algorithm of the proposed channel allocation scheme is as follows .
using the algorithm described above , the minimum number of distinct channels required without any overlap of the coverage areas for four zones are given in fig . [ channel ] . under this channel allocation scheme ,
the maximum number of distinct tv channels required in the entire zone is four , which is much smaller than the fourteen channels currently used in india . to avoid adjacent channel interference
, the overlapping channels must be non - adjacent .
in this paper , quantitative analysis of the available tv white space in the @xmath0-@xmath1mhz uhf tv band in india was performed .
it is observed that unlike developed countries , a major portion of tv band spectrum is unutilized in india .
the results show that even while using conservative parameters , in at least @xmath72% areas in the four zones all the @xmath42 channels ( @xmath2% of the tv band spectrum ) are free !
the average available tv white space was calculated using two methods : ( i ) the protection and pollution viewpoints @xcite ; and , ( ii ) the fcc regulations @xcite . by both methods , the average available tv white space in the uhf tv band was shown to be more than @xmath2mhz !
an algorithm was proposed for reassignment of tv transmitter frequencies to maximize unused spectrum .
it was observed that four tv channels ( or @xmath3mhz ) are sufficient to provide the existing uhf tv band coverage in india . in the future ,
we plan to obtain and include the missing north zone data in our work .
we also wish to explore suitable regulations in india for the tv white space to enable affordable broadband coverage .
this is timely and important since policy intent for tv white space was made in nfap 2011 .
m. islam et .
al , spectrum survey in singapore : occupancy measurements and analyses , " in proc . of 3rd intl .
conference on cognitive radio oriented wireless networks and communications , may 2008 , pp . 1 - 7 .
m. mchenry et .
chicago spectrum occupancy measurements & analysis and a long - term studies proposal , " in proc . of the acm 1st intl .
wkshp . on tech . and
policy for accessing spectrum , aug .
2006 , pp . 1 - 12
. t. shimomura , t. oyama and h. seki , `` analysis of tv white space availability in japan , '' in proc . of ieee vehicular tech .
p. kumar et .
al , white space detection and spectrum characterization in urban and rural india , " in proc . of ieee 14th
symp . and wkshops on a world of wireless , mobile and multimedia networks , jun .
2013 , pp . 1 - 6
. ofcom , `` digital dividend : cognitive access .
consultation on license - exempting cognitive devices using interleaved spectrum , '' feb . | licensed but unutilized television ( tv ) band spectrum is called as tv white space in the literature .
ultra high frequency ( uhf ) tv band spectrum has very good wireless radio propagation characteristics .
the amount of tv white space in the uhf tv band in india is of interest .
comprehensive quantitative assessment and estimates for the tv white space in the @xmath0-@xmath1mhz band for four zones of india ( all except north ) are presented in this work .
this is the first effort in india to estimate tv white spaces in a comprehensive manner .
the average available tv white space per unit area in these four zones is calculated using two methods : ( i ) the primary ( licensed ) user and secondary ( unlicensed ) user point of view ; and , ( ii ) the regulations of federal communications commission in the united states . by both methods ,
the average available tv white space in the uhf tv band is shown to be more than @xmath2mhz ! a tv transmitter frequency - reassignment algorithm is also described .
based on spatial - reuse ideas , a tv channel allocation scheme is presented which results in insignicant interference to the tv receivers while using the least number of tv channels for transmission across the four zones . based on this reassignment , it is found that four tv band channels ( or @xmath3mhz ) are sufficient to provide the existing uhf tv band coverage in india . |
the physical picture of our theory for the forced dissociation of receptor - ligand bonds is very similar with the small ligand binding to heme proteins @xcite : there is a energy surface for dissociation which dependents on both the reaction coordinate for the dissociation and the conformational coordinate @xmath7 of the complex , while the later is perpendicular to the former ; for each conformation @xmath7 there is a different dissociation rate constant which obeys the bell rate model , while the distribution of @xmath7 could be modulated by the force component along x - direction ; higher temperature or larger diffusivity ( low viscosities ) allows @xmath7 variation within the complex to take place , which results in a variation of the energy barrier of the bond with time .
there are two types of experimental setups to measure forced dissociation of receptor - ligand complexes .
first we consider constant force mode @xcite .
a diffusion equation in the presence of a coordinate dependent reaction is given by @xcite @xmath8 where @xmath9 is probability density for finding a value @xmath7 at time @xmath10 , and @xmath11 is the diffusion constant .
the motion is under influence of a force modulating potential @xmath12 , where @xmath13 is intrinsic potential in the absence of any force , and a coordinate - dependent bell rate . in the present work
[ bellmodel ] depends on @xmath7 through the intrinsic rate @xmath14 , and the distance @xmath15 is assumed to be a constant for simplicity . here
@xmath16 and @xmath17 are respective projections of external force @xmath4 along the reaction and conformational diffusion coordinates : @xmath18 and @xmath19 is the angle between @xmath4 and the reaction coordinate .
we are not ready to study general potentials here .
instead , we focus on specific @xmath13s , which make @xmath20 to be @xmath21 where @xmath22 and @xmath23 are two constants with length and force dimensions .
for example for a harmonic potential @xmath24 with a spring constant @xmath25 in which we are interested , it gives @xmath26 and @xmath27 defining a new coordinate variable @xmath28 , we can rewrite eq .
[ origindiffusionequation ] with the specific potentials into @xmath29 where @xmath30 .
compared to the original work by agmon and hopfield @xcite , our problem for the constant force case is almost same except the reaction rate now is a function of the force .
hence , all results obtained previously could be inherited with minor modifications .
considering the requirement of extension of eq .
[ origindiffusionequation ] to dynamic force in the following , we present the essential definitions and calculations .
substituting @xmath31 into eq .
[ newdiffusionequation ] , one can convert the diffusion - reaction equation into schr@xmath32dinger - like presentation @xcite .
@xmath33 where @xmath34 is the normalization constant of the density function at @xmath35 , and the effective " potential @xmath36+k_f(y).\nonumber\end{aligned}\ ] ] we define @xmath37 for it is independent of the force @xmath4 .
. [ schodingerequation ] can be solved by eigenvalue technique @xcite . at larger @xmath11 in which we are interested here
, only the smallest eigenvalue @xmath38 mainly contributes to the eigenvalue expansion which is obtained by perturbation approach @xcite : if the eigenfunctions and eigenvalues of the unperturbed " schr@xmath32dinger operator @xmath39 in the absence of @xmath40 have been known , @xmath41 and @xmath42 is adequately small , the first eigenfunction @xmath43 and eigenvalue @xmath38 of the operator @xmath44 then are respectively given by @xmath45 and @xmath46 considering that the system is in equilibrium at the initial time , _
i.e. _ , no reactions at the beginning , the first eigenvalue @xmath47 must vanish . on the other hand , because @xmath48 and the square of @xmath49 is just the equilibrium boltzmann distribution @xmath50 with the potential @xmath51 , we rewritten the first correction of @xmath38 as @xmath52.\nonumber\end{aligned}\ ] ] substituting the above formulaes into eq .
[ transform ] , the probability density function then is approximated to @xmath53\phi_0(f)\end{aligned}\ ] ] the quantity measured in the constant force experiments is the mean lifetime of the bond @xmath54 , @xmath55 where the survival probability @xmath56 related to the probability density function is given by @xmath57.\end{aligned}\ ] ] in addition to the constant force mode , force could be time - dependent , _ e.g. _ , force increasing with a constant loading rate in biomembrane force probe ( bfp ) experiment @xcite . in principle
the scenario would be more complicated than that for the constant force mode .
we assume that the force is loaded slowly compared to diffusion - reaction process .
we then make use an adiabatic approximation analogous to what is done in quantum mechanics .
the correction of this assumption would be tested by the agreement between theoretical calculation and experimental data .
we still use eq .
[ origindiffusionequation ] to describe bond dissociations with the dynamic force , therefore we obtain the almost same eqs . [ forcedependentpotential]-[forcedependentquantumpotential ] except that the force therein is replaced by a time - dependent function @xmath58 .
we immediately have @xcite @xmath59\phi_0(f_t),\end{aligned}\ ] ] where the berry phase " @xmath60 and @xmath61 is the first eigenfunction of the time - dependent sch@xmath32dinger operator @xmath62 because the eigenvalues and eigenfunctions of the above operator can not be solved analytically for general @xmath63 , we also apply the perturbation approach .
hence , we obtain @xmath61 and @xmath64 by replacing @xmath42 in eqs .
[ eigenfunctionexpansion ] and [ eigenvalueexpansion ] with @xmath63 .
the berry phase then is approximated to @xmath65 finally , the survival probability for the dynamic force is given by @xmath66\nonumber\\\end{aligned}\ ] ] different from the constant force mode , data of the dynamic force experiments is typically presented in terms of the force histogram , which corresponds to the probability density of the dissociation forces @xmath67 @xmath68 particularly , when the force is a linear function of time @xmath69 , where @xmath70 is the loading rate , and zero or nonzero of @xmath71 respectively corresponds to the steady- or jump - ramp force mode in the dynamic force experiment @xcite , we have @xmath72\times\\ & & \exp\left[-\frac{1}{r}\int_{f_0}^f\left(\lambda^{(1)}_0(f')+ \lambda^{(2)}_0(f')+ b(f')\right)df'\right].\nonumber\end{aligned}\ ] ]
we consider a bounded diffusion in the harmonic potential eq . [ harmonicpotential ] .
then @xmath73 reduces to a harmonic oscillator operator with @xmath74 its eigenvalues and eigenfunctions are @xmath75 and @xmath76 respectively , where @xmath77 and @xmath78 is the hermite polynormials @xcite .
given that the intrinsic dissociation rate satisfies the arrenhenius form @xmath79,\end{aligned}\ ] ] where the height of the energy barrier along the reaction coordinate @xmath80 is a function of the conformational coordinate @xmath7 .
according to the form of barrier , we first analyze two simple and meaningful cases .
+ * bell - like forced dissociations*. the simplest function of the energy barrier might be linear with respect to @xmath7 , @xmath81 where @xmath82 is the height at position @xmath83 , and the slope @xmath84 ( its dimension force ) for the perturbation requirement in solving eq .
[ schodingerequation ] . according to eqs .
[ eigenvalueexpansion ] and [ berryphase ] , we easily get @xmath85\times\nonumber\\ & & \exp\left[\beta\left(\xi^\ddag f_\parallel-\frac{k_g}{k_x}f_\perp\right)\right],\nonumber\\ \lambda^{(2)}_0(f)&=&\frac{-k_0 ^ 2}{\beta dk_x}\exp{\left[-2\beta\delta g^\ddag_0+\frac{\beta
k_g^2}{k_x}\right]}\times\\ & & \exp{\left[2\beta\left(\xi^\ddag f_\parallel-\frac { k_g}{k_x}f_\perp\right)\right]}\sum_{n=1}^{\infty}\frac{1}{nn!}\left(\frac{\beta k_g^2}{k_x}\right)^n,\nonumber\end{aligned}\ ] ] and @xmath86}\times\\ & & \exp{\left[2\beta\left(\xi^\ddag { f_t}_\parallel-\frac { k_g}{k_x}{f_t}_\perp\right)\right ] } \sum_{n=1}^{\infty}\frac{1}{n^2n!}\left(\frac{\beta k_g^2}{k_x}\right)^n,\nonumber\end{aligned}\ ] ] where @xmath87 .
for large @xmath11 or @xmath25 ( or very small @xmath6 ) , the second correctness and the berry phase tend to zero . under these limitations
the first eigenvalue of eq .
[ origindiffusionequation ] is approximated to be @xmath88
\exp[\beta d^\ddag f].\end{aligned}\ ] ] here we define a new distance @xmath89 where @xmath90 whose dimension is distance .
we see that the presence of the complex conformational coordinate could modify the original bell model in novel ways : ( i ) @xmath91 , eq .
[ correctedbellform ] is indistinguishable from the origin bell model , although the projection distance @xmath92 from the bound state to the energy barrier may be increased or decreased in terms of the orientation of the applied force .
in particular , if the force is antiparallel to @xmath7 , _
i.e. _ , @xmath93 , we get a bell - like rate expression with a distance " @xmath94 ; ( ii ) @xmath95 , the force does not affect dissociations of the bonds , which have been named ideal " bonds by dembo @xcite ; ( iii ) @xmath96 , the force slows down dissociations of the bonds .
it is catch " bonds in which we are interested .
in contrast to the catch behavior suggested by dembo @xcite , the rate decays exponentially with respect to the force instead of the square of the force .
given the linear function eq .
[ linearheightfunction ] and @xmath97 , increasing of the force only stabilizes the bonds by dragging the system to the higher energy barriers ( catch ) , whereas the other force component @xmath98 destabilizes the complex by lowering the energy barriers ( slip ) .
therefore the sign of the distance @xmath99 in fact reflects a competition of the two contrast effects of the same force . + * dembo - like forced dissociations*. another function of the energy barrier is a harmonic with a spring constant @xmath100 @xmath101 where @xmath102 is the barrier height at position @xmath103 . because for any form of the barrier height , the dependence of @xmath104 and @xmath105 on @xmath11 is the same from eq .
[ unpertubatedeigenvaluesharmonic ] , we only consider the large @xmath11 limitation in the following .
hence we have @xmath106\times\nonumber\\ \exp\left[-\frac{\beta k_g(f_\perp - k_x(x_1-x_0))^2}{2k_x(k_x+k_g)}\right]\end{aligned}\ ] ] given @xmath107 and @xmath108 , we find that there is a interesting transition from slip to catch bond when the force increase over a threshold @xmath109 ; otherwise only catch bond presents .
we note that the latter is very similar to the result proposed by dembo @xcite even their physical origins are completely different : both of them exponentially dependent on the square of the force . + * comparison with the experiments . * . @xmath110 and @xmath111 are the value of the linear functions in eq .
[ bendfunction ] at position @xmath83 , while @xmath112 is the intersection of the functions at @xmath113 .
the bold solid line is the minimum model which is used to fit the experiment .
the shaded area represents the equilibrium distribution of the conformational coordinate under the potential @xmath51 ( eq . [ equilibriumdistribution ] ) . ] in the constant force rupture experiment of the psgl-1@xmath0p - selectin complex , the dissociation rate as the inverse mean lifetime of the complex first decreased and then increased when the applied force increased beyond a force threshold @xcite .
we now can easily understand this counterintuitive transition according to the previous discussion about the bell - like dissociation rate : the dissociation effect of @xmath98 regains its dominance when the force is beyond the threshold .
although in principle we can construct various barrier height functions which result into catch - slip transitions , the most simplest form may be a composition of two linear functions @xmath114 where we require that the distances defined in eq .
[ newdistance ] with @xmath115 and @xmath116 are respectively minus and positive . for convenience , their absolute values
are correspondingly denoted by @xmath117 and @xmath118 .
define two intrinsic " dissociation constants @xmath119,\nonumber\\ k^s_0=k_0\exp[-\beta\delta g^\ddag_s(x_0)].\end{aligned}\ ] ] fig . [ figure1 ] shows the characteristics of the function .
we then have @xmath120\exp\left[-\beta d^\ddag_c f\right]\nonumber\\ & & \times{\rm erfc}\left[-\left(\delta+\frac{k_c}{k_x}\right)\sqrt{\frac{\beta k_x}{2}}+f\sqrt{\frac{\beta}{2k_x}}\sin\theta \right ] \nonumber\\ & & + \text { } \frac{k_0^s}{2}\exp\left[\frac{\beta
k_s^2}{2k_x}\right]\exp\left[\beta d^\ddag_s f\right]\\ & & \times{\rm erfc}\left[\left(\delta+\frac{k_s}{k_x}\right)\sqrt{\frac{\beta k_x}{2}}-f\sqrt{\frac{\beta}{2k_x}}\sin\theta \right]\nonumber\end{aligned}\ ] ] where @xmath121 , and the complementary error function @xmath122 before fixing numerical values of the parameters in eq .
[ reactionratebendlandscape ] , we first simply analyze the main properties of @xmath38 given @xmath123 : ( i ) in the absence of force , due to @xmath124 and @xmath125 , we have @xmath126 which is the same with that obtained by agmon and hopfield @xcite ; ( ii ) if force is nonzero and smaller , @xmath127 which means that the bond is catch ; and finally ( iii ) , when the force is sufficiently large , eq .
[ reactionratebendlandscape ] reduces to @xmath128 it is the ordinary slip bond .
there are total eight independent parameters presenting in eq .
[ reactionratebendlandscape ] : @xmath129 , and @xmath130 .
it is not necessary to determine all of them , which is also impossible only through fitting to the experimental data @xcite .
for example , the latter two parameters are lumped into @xmath131 , while and @xmath15 always presents with @xmath132 together .
what we really concern with is the coefficients of the force and the factors before the error functions in eq .
[ reactionratebendlandscape ] .
they can be obtained by least square fit .
guided by the properties eqs [ smallforcelimit ] and [ largeforcelimit ] , the fitting process in fact is simple .
even so , we are still able to fix all parameters from the fitting results if we study a minimum model in which the slop @xmath116 is zero and @xmath133 . here
the particular value of the angle is actually of no particular significance and it is only as a reference .
we immediately have : @xmath134 , @xmath135 , @xmath136 , @xmath137 ; the other interesting parameters see tab .
1 , where the values are independent of the angle . .comparison of the parameters of the present theory , and the two - pathway and one energy - well model presented by thomas et al . @xcite on the constant force ( cf ) and the dynamic force ( df ) modes .
the parameters for the slip behavior of the p - selectin are also listed as a reference @xcite . [ cols="^,^,^,^,^,^",options="header " , ] [ table ] substituting these values into eq .
[ reactionratebendlandscape ] and according to eq .
[ averagelifetime ] , we calculate the mean lifetime of the psgl-1@xmath0p - selectin complex with respect to different constant force in fig .
[ figure2 ] : the agreement between theory and the experimental data is quite good . with the same parameters . ]
more challenging experiments to our theory are the force steady- and jump - ramp modes @xcite . under large @xmath11 limitation , eq . [ ruputureforcedistribution ] reduces to @xmath138 . \label{approxruputureforcedistribution}\end{aligned}\ ] ] we see that the mean lifetime can be extracted from the above equation by setting @xmath139 , _
i.e. _ , @xmath140 .
we calculate the dissociation force distributions of the steady- and jump - ramp modes at three loading rates to compare with the bfp experiments performed by evans _ et al . _
here we are not ready to fit the experiments afresh ; instead we directly apply the parameters obtained from the constant force mode to current case . because the bfp experimental data is for _ dimeric _ ligand psgl-1 , whereas our parameters are from monomeric ligand spsgl-1 .
therefore it is necessary to map our predictions for the single bond spsgl-1@xmath0p - selectin to the double bonds .
a natural assumption is that the two bonds share the same force and fail randomly .
the same assumption has been used in previous works @xcite .
hence , the probability density of the dissociation force for the double bond psgl-1@xmath0p - selectin complex is related to the single case by @xmath141 fig .
[ figure3 ] presents the final result .
we see that the theoretical prediction agrees to the data very well .
hence we conclude that the adiabatic approximation proposed at the beginning is reasonable . under the different loading rates predicted by our theory ( solid curves ) for the psgl-1@xmath0p - selectin complex .
the symbols are from the force steady- and jump - ramp experimental data @xcite .
the apparent deviations between the theory and the data in the last column may be from the invalidation of the assumption of two equivalent bonds at higher loading rates . ]
the previous works @xcite have claimed that they could not fit the experimental data from the constant force experiment using atomic force microscopy ( afm ) and the force jump - ramp experiment using bfp with the same parameters , _
e.g. _ , see tab .
the authors simply contributed it to the different equipment and biological constructs though the experiments studied the same complexes @xcite .
our calculations however show that the mechanical parameters defined by us have almost the same values . in addition , the tendencies of our density functions for the first two panels of the second array in fig .
[ figure3 ] are closer to the data than that predicted by the two - pathways models @xcite .
the density functions and the force histograms in the experiments reach the maximum and minimum at two distinct forces , which are named @xmath142 and @xmath143 in the following , respectively .
this observation could be understood by setting the derivative of eq .
[ ruputureforcedistribution ] with respect to @xmath4 equal to zero , @xmath144 we immediately see that the values of @xmath142 and @xmath145 must be larger than the catch - slip transition force @xmath146 for the left term in eq .
[ extremacondition ] is negative as the bond is catch . indeed
the experimental observations show that the force values at the minimum histograms are around a certain values even the loading rates change 10-fold .
( see figs . 2 and 4 in ref .
the above equation could have no solutions when the loading rate is smaller than a critical rate @xmath147 , which can be obtained by simultaneously solving eq .
[ extremacondition ] and its first derivative .
we estimate @xmath148 pn / s using the current parameters , while the force @xmath149 pn ( about 26 pn in the double bond cases ) . if @xmath150 , then the density function is monotonous and decreasing function .
therefore , the most probable force at the bond dissociation is zero .
the most interesting characteristics of eq .
[ extremacondition ] are the dependence of the maximum and minimum forces on the loading rate .
in particular the latter is an important index in dynamic force spectroscopy ( dfs ) theory since it corresponds to the most possible dissociation force @xcite . when the loading rate is sufficiently large , and correspondingly @xmath143 is larger , the approximation of eq . [ reactionratebendlandscape ] at large force eq .
[ largeforcelimit ] implies that @xmath151}\propto \ln r.\end{aligned}\ ] ] the experimental measurement supported this prediction ; see fig .
3a in ref .
@xcite . on the other hand , due to
that @xmath142 is very close to @xmath146 at the larger loading rate , employing the taylor s expanding approach we have @xmath152,\end{aligned}\ ] ] where @xmath153 . substituting it into eq .
[ extremacondition ] , we get @xmath154 it means that @xmath142 tends to @xmath146 very fast .
different from @xmath143 , the loading rate dependence of @xmath142 is an intrinsic property of the catch - slip bond ; a unique requirement is that the dissociation rate @xmath38 has a minimum at the transition force @xmath146
. therefore @xmath142s observed in experiment performed by evans _
et al . _
@xcite are almost the catch - slip transition force observed in the constant force rupture experiment performed by marshall _
_ @xcite . indeed , the force values of the minimum force histograms for the former are about 26 pn , while the transition force for the latter ( _ dimeric _ psgl-1@xmath0p - selectin ) is also about 26 pn . we know that the dissociation forces distribution of a simple slip bond only has a maximum at a certain force value that depends on loading rate @xcite . therefor the catch - slip bond
can easily be distinguished from the slip case by the presence of a minimum on the density function of the dissociation forces at a nonvanished force .
because the above analysis is independent of the initial force @xmath71 , in order to track the catch behaviors in the force jump - ramp experiments , @xmath71 should be chosen to be smaller than @xmath146 .
compared to the chemical kinetic schemes , our theory should be more attractive on the following aspects .
first of all , we suggest that the counterintuitive catch - slip transition is a typical example of the rate processes with dynamic disorder .
because this concept has been broadly and deeply studied from theory and experiment during the past two decades , extensive experience and knowledge could be used for reference .
for example , we suggest that a new receptor - ligand forced dissociation experiment could be performed over a large range of temperatures and solvent viscosities . according to eq .
[ origindiffusionequation ] , if the viscosity is so higher that @xmath155 , we could predict @xmath156.\end{aligned}\ ] ] we know that such a dissociation reaction is a typical example of the rate processes with _ static _ disorder @xcite .
in addition that the survival probability of the bond converts into multiple exponential decay at a single force from the single exponential decay at the large @xmath11 limitation ( see eq . [ survivalprobability ] ) , the mean lifetime is @xmath157 which means that the catch - slip bond changes into slip bond only .
then our theory gives a intuitively obvious physical explanation of catch bonds in an apparent expression ( eq .
[ correctedbellform ] ) : they could arise from a competition of the two components of applied external force along the dissociation reaction coordinate and the molecular conformational coordinate ; the former accelerates the dissociation by lowering the height of the energy barrier , while the latter stabilizes the complex by dragging the system to the higher barrier height . finally , the time - dependence of the forced dissociation rates could be induced by either global conformational changes of the complex or local conformational changes at the interface between the receptor and ligand ; no separated bound states and pathways are needed in the current theory .
therefore it is possible that one can not find new stable complex structures through experiments or detailed molecular dynamics ( md ) simulations .
even there are many advantages in the present theory .
we can not definitely distinguish which theory or model is the most reasonable and more close real situations with existing experimental data .
moreover , except the coarse - grain physical picture our theory does not reveal the detailed structural information of the catch behavior of the ligand - receptor complexes , while biologists might be more interested in it .
we could correspond the increasing height of the energy barrier with respect to the conformational coordinate to the hook structure @xcite or more affinity bound states @xcite , however we believe that further single - molecule experiments including micromanipulation experiments and fluorescence spectroscopy , more crystal structure data and detailed md simulations from the atomic interactions are essential to elucidate the real molecular mechanism of the catch bonds .
+ + fl thanks prof .
mian long and dr .
fei ye for their helpful discussion about the work . | recently experiments showed that some adhesive receptor - ligand complexes increase their lifetimes when they are stretched by mechanical force , while the force increase beyond some thresholds their lifetimes decrease .
several specific chemical kinetic models have been developed to explain the intriguing transitions from the catch - bonds " to the slip - bonds " . in this work
we suggest that the counterintuitive forced dissociation of the complexes is a typical rate process with dynamic disorder . an uniform one - dimension force modulating agmon - hopfield model is used to quantitatively describe the transitions observed in the single bond p - selctin glycoprotein ligand 1(psgl-1)@xmath0p - selectin forced dissociation experiments , which were respectively carried out on the constant force [ marshall , _
et al . _ , ( 2003 ) nature * 423 * , 190 - 193 ] and the force steady- or jump - ramp [ evans _ et al . _ , ( 2004 ) proc .
natl .
acad .
sci .
usa * 98 * , 11281 - 11286 ] modes .
our calculation shows that the novel catch - slip bond transition arises from a competition of the two components of external applied force along the dissociation reaction coordinate and the complex conformational coordinate : the former accelerates the dissociation by lowering the height of the energy barrier between the bound and free states ( slip ) , while the later stabilizes the complex by dragging the system to the higher barrier height ( catch ) .
adhesive receptor - ligand complexes with unique kinetic and mechanical properties paly key roles in cell aggregation , adhesion and other life s functions in cells .
a well studied example is the receptors in selectin family which comprises e- , l- and p - selectin interacting and forming bonds " with their glycoprotein ligands .
these bonds are primarily responsible for the tethering and rolling of leukocytes on inflamed endothelium under shear stress @xcite . in particular , in the past two years great experimental efforts @xcite have been devoted to study the surprising kinetic and mechanical behaviors of the bonds between l- and p - selectin and p - selectin glycoprotein ligand 1 ( psgl-1 ) at the single molecule level : the lifetimes of these bonds first increase with initial application of small force , which are termed catch " bonds , and subsequently decrease , which are termed slip " bonds when the force increases beyond some thresholds .
the most important biological meaning of this discovery is that the catch - slip transitions of the psgl-1@xmath0l- and @xmath0p - selectin bonds may provide a direct experimental evidence at the single - molecule level to account for the shear threshold effect @xcite , in which the number of rolling leukocytes first increases and then decreases while monotonically increasing shear stress . on the theoretical side , it is a challenge to give a reasonable physical theory or model to explain the counterintuitive bond transitions .
bell @xcite firstly suggested that the force induced dissociation rate of adhesive receptor - ligand complex could be described by , @xmath1 , \label{bellmodel}\end{aligned}\ ] ] where @xmath2 is the intrinsic dissociation rate constant in the absence of force , @xmath3 is the distance from the bound state to the energy barrier , @xmath4 is a projection of external applied force along the dissociation coordinate , @xmath5 the boltzmann s constant , and @xmath6 is absolute temperature .
the validity of the model has been demonstrated in experiments @xcite . although later at least four models have been put forward to explain and understand various receptor - ligand forced dissociation experiments @xcite , they can not predict catch bonds because force in these models only lowers height of the energy barrier while shortening lifetimes of the bonds .
an exception is the hookean spring model proposed by dembo many years ago @xcite , in which a catch bond was raised in mathematics .
compared to the exponential decay of the lifetimes of slip - bonds with respect to force in experiments @xcite , the model claimed that the lifetimes decrease exponentially with the square of force @xcite .
in addition , the hookean spring model can not account for the catch - slip bond transitions in self - consistent term .
prompted by the intriguing experimental observations , three chemical kinetic models have been developed . evans _
et al . _
presented a two pathways , two bound states model with rapid equilibrium assumption between the two states .
they suggested that the catch - slip bond transitions take place due to applied force switching the pathways from the one with slower dissociation rate to the fast one @xcite .
although this insightful viewpoint well described force jump - ramp experiments , there are two apparent flaws in physics .
first if the forced dissociation experiments were performed at very low temperatures or higher solvent viscosities , force would be independent of the bond dissociations since the force in the two pathways and two bound states model only acts on the inner bound states , while these states would be frozen " under this circumstances .
the other is that force does not accelerate the dissociation processes further when the force is sufficiently large for the fast dissociation rate is a constant .
the next model given by barsegov and thirumalai @xcite with same kinetic scheme seems to improve the two flaws in which the dissociation rates of the two pathways were allowed to be force - dependent with the bell formulas .
unfortunately , so many independent reaction constants with arbitrary dependencies on the force parameters ( total seven parameters ) and the final dissociation rate depending on them in a complicated way make the physical explanations and the determination of the parameters difficult to track . very recently , a competitive two pathways and one bound state model was proposed by thomas _
et al . _
@xcite .
this model is distinct from the others because there is a catch pathway therein , which was thought to arise from a backward unbinding pathway .
but the model is not intuitively obvious just like the authors pointed out . as one type of noncovalent bonds , interactions of adhesive receptors and their ligands
are weaker .
moreover , the interfaces between them have been reported to be broad and shallow , such as the crystal structure of the psgl-1@xmath0p - selectin bond revealed @xcite .
therefore it is plausible that the energy barriers for the bonds are fluctuating with time due to either global conformational changes or local conformational changes at the interfaces .
association / dissociation reactions with fluctuating energy barrier have been deeply studied by statistical physicists in terms of rate processes with _ dynamic disorder _
@xcite during the past two decades .
a prototype is the ligand rebinding in myoglobin where the rate constant depends on a protein coordinate @xcite .
hence it is of interest to determine whether the fluctuation of energy barrier responses to catch - slip bond transitions .
such studies should be meaningful since in the bell s initial work and the other models developed later , the intrinsic rate constants @xmath2 were deterministic and time - independent .
it is possible to derive unexpected results from the relaxation of this restriction .
stimulated by the two considerations , in the present work we propose that the intrinsic dissociation rate in the bell model is controlled by a conformational coordinate of receptor - ligand complex , while the coordinate is fluctuating as a brownian motion in a bound harmonic potential ; applied force not only lowers the height of the energy barrier as described in eq . [ bellmodel ] but also modulates the distribution of the conformational coordinate .
in addition to well predicting the experimental data , our theory may also provide a new physical mechanism for the dissociation rates suggested by bell @xcite and dembo @xcite early . |
the measurement of the extragalactic background light ( ebl ) is important for vhe gamma - ray astronomy as well as for astronomers modeling star formation and galaxy evolution .
second only in intensity to the cosmic microwave background , the optical and infrared ( ir ) ebl contains the imprint of galaxy evolution since the big bang .
this includes the light produced during formation and re - processing of stars .
current measurements of the ebl are summarized in fig . [
fig : ebl_ir ] .
the optical to near - ir emission , peaking in the 1 @xmath2 m region and extending to 20 @xmath2 m is due to direct star light , while molecular clouds and dust reprocess the optical light and emit in the mid to far ir region producing the second peak , @xmath3 20 - 300 @xmath2 m . @xcite comprehensively reviewed measurements and implications of the cosmic infrared background .
the optical to far - infrared ebl is difficult to measure because it is dwarfed by the much brighter foregrounds caused by night - sky glow , diffuse dust in the galaxy , and the zodiacal light caused by interplanetary dust @xcite .
for example , emission by the zodiacal dusk peaks in the 25 @xmath2 m region , orders of magnitude above the low ebl density in this waveband . in the case of ground- or rocket - based observations ,
instrumental emission also plays a significant role .
this is complicated by the fact that the only characteristic upon which a detection of the ebl can be based is that it has to be distributed isotropically .
these difficulties have precluded ground- and rocket - based measurements from detecting the ebl at all @xcite .
the measured flux of vhe gamma rays is attenuated by pair production with optical / ir photons .
on extragalactic distances , most of this absorption occurs by the ebl @xcite ; interactions with stellar and galactic material and optical / ir photons are negligible @xcite .
thus , if one somehow knows the initial gamma - ray flux , vhe astronomy is in a unique position to place limits on the density of the intervening optical / ir component of the ebl .
the organization of this paper is as follows : existing measurements and constraints on the ebl are presented in sect .
[ sec : ebl_measurements ] .
then , the brightest flare spectra from each of the six blazars are presented in sect .
[ sec : comparison_of_spectra ] together with the apparent spectral steepening with redshift .
this is followed by a short review of pair - production absorption in sect .
[ sec : theory ] and is illustrated for the particular case of an assumed monoenergetic ebl in sect .
[ sec : monoenergetic ] .
upper limits on the ebl density are derived in sect .
[ sec : upper_limits ] and conclusions are given in sect . [
sec : conclusion ] .
direct measurements are possible in the two windows of least foreground around 1 @xmath2 m and @xmath0100 @xmath2 m @xcite .
recently , the cosmic background explorer ( cobe ) satellite with its two instruments , the diffuse infrared background experiment ( dirbe ) and the far infrared spectrometer ( firas ) , has detected the ebl at 140 @xmath2 m and 240 @xmath2 m , see fig .
[ fig : ebl_ir ] . the possible detections at 60 @xmath2 m and 100 @xmath2 m @xcite are viewed as too high and are controversial , requiring revised galaxy evolution models with larger dust content @xcite .
the firas measurement @xcite shows that the ir ebl can be characterized between 125 @xmath2 m and 2000 mm by a modified blackbody spectrum .
the isotropic optical and near - ir emission detected with the japanese irts satellite is considerably higher than integrated light from galaxies and theoretical predictions @xcite .
lower limits on the ebl density are placed by adding the flux per unit area received from all galaxies down to a given flux limit .
as galaxies are only one source contributing to the ebl , these galaxy counts represent a lower limit on the total ebl @xcite . in the mid - ir region , where the foreground is particularly bright , @xcite were able to place a lower limit on the 15 @xmath2 m ebl density .
upper limits can be placed on the ebl from direct measurements with minimal background subtraction @xcite .
also , upper limits can be derived from fluctuations in the measured light distribution , see @xcite .
bl lacertae ( bl lac ) objects are one type of active galactic nuclei that have a jet of high energy particles aligned with our line of sight .
the nonthermal emission from the jet is highly variable and the mechanism for production of the vhe gamma - ray peak is under debate . in all models
the intrinsic vhe spectrum of the source is smooth and concave downwards ; no physical mechanism has been proposed that would produce an exponential rise with energy or emission - like line features . the vhe spectrum measured on earth
is modified by interactions of gamma rays with the ebl .
for example , a 1 tev gamma ray reaches the threshold for electron - positron pair production if it collides with a 1 ev optical photon .
if a pair is created , this gamma ray is lost and causes an attenuation of the measured spectrum at that energy .
thus , the ebl density could be inferred from the measured spectrum if the intrinsic spectrum of the blazar were known .
in addition , because of the coarse energy resolution of this method , a measurement of the ebl spectrum is only possible if the shape of the ebl spectrum is known in advance .
three methods have been used so far to derive ebl limits : 1 . using multiwavelength observations ,
the intrinsic vhe blazar spectrum is predicted through interpolation or fitting of an emission model .
the ebl spectral shape is usually fixed in advance with only the absolute normalization of the ebl flux left as a free parameter @xcite .
the deviation , or lack thereof , of the vhe spectrum from a predefined intrinsic spectrum , e.g. a power - law , is used to place upper limits on the ebl density with a given spectral shape @xcite .
3 . for a given ebl scenario
the intrinsic source spectrum is reconstructed .
ebl models are ruled out , or limits are placed on the density , if they give rise to an unphysical source spectrum : such as that it must not rise exponentially and that it should be consistent with the x - ray synchrotron peak @xcite .
the method described in this paper differs from the above in that : ( 1 ) the intrinsic source spectrum of all detected vhe blazars is assumed to be identical to each other .
a conservative upper limit to the intrinsic spectrum is interpolated from the observed correlation between the spectral index of the source during a flare state and its redshift .
( 2 ) upper limits to the ebl intensity are calculated only at single energy points where the vhe spectrum has been measured , avoiding the need to specify the ebl spectral shape .
vhe emission has been detected and confirmed from six bl lac objects : mrk 421 @xcite , mrk 501 @xcite , 1es 2344 + 514 @xcite , 1es 1959 + 650 @xcite , pks 2155 - 304 @xcite , and h 1426 + 428 @xcite .
for all six blazars , hysteresis and the rarity of simultaneous x - ray / vhe spectra make it difficult to establish a temporal correlation between x - rays and vhe gamma rays .
the sources are similar in that the x - ray synchrotron peak ranges from @xmath4 to almost @xmath5 hz with peak luminosities between @xmath6 and @xmath7 erg s@xmath8 sr@xmath8 @xcite . the black hole mass estimates are also relatively similar : @xmath9-@xmath10 m@xmath11 .
it is possible that the reason for this similarity lies in the fact that only the brightest vhe blazars are seen by the imaging atmospheric cherenkov telescopes at this time and that there is actually a continuous population of blazar spectra @xcite .
the brightest flare spectra measured from each of the six blazars are shown in fig .
[ fig : spectra_agn ] ; they are increasingly steep with source distance .
the spectra for mrk 421 @xcite and mrk 501 @xcite are best described by a power law with exponential cut - off @xcite .
the cut - off , at energy of about 4 tev , can be attributed to gamma - ray absorption by the extragalactic medium . for 1es 1959 + 650 ,
the gamma - ray orphan flare is used @xcite , a tev flare during which the x - ray spectrum was not particularly bright @xcite . for 1es 2344 + 514 , the brightest flare recorded thus far was recorded in 1995 by the whipple 10 m telescope ;
the spectrum has been derived in @xcite .
the spectrum of pks 2155 - 304 has recently been measured by the hess collaboration with high precision @xcite .
though the light curve shows variability , the spectrum can be well fitted by a single power law over the entire energy range while the power law index did not change with flux level .
vhe emission from the furthest object , h 1426 , is generally very low @xcite , so that no flare spectra have been measured .
however , as vhe emission during flaring episodes is much higher than during the quiescent state , most of the photons in the spectrum by @xcite may be associated with flares .
the power - law spectral index fitted to each agn flare spectrum is shown in fig .
[ fig : z_index ] . because a significant cutoff is present in the spectra of mrk 421 and mrk 501 ,
the spectral index is measured at two energies , 1 and 2 tev , to estimate the systematic error arising from this choice . in both cases , the increasingly steep spectral index with redshift
was fitted with a straight line , resulting in a spectral index at @xmath12 of [email protected] with @xmath14/(d.o.f . )
= 22.3/4 when the spectral index is measured at 1 tev and index of [email protected] ( @xmath14/(d.o.f.)=5.7/4 ) when the index is measured at 2 tev .
the flare spectra , after taking into account the redshift dependence , appear similar and hence we make the zeroth order assumption that the same process is responsible for the vhe emission in all these agn
. in the interpretation that the cut - off in the energy spectrum of mrk 421 and mrk 501 is due to absorption by the ebl , note that for a power law fit with exponential cut - off , the power law index ranges from -1.9 to -2.1 , with a cut - off energy around 4 tev consistent for both blazars and independent of flux state @xcite .
the literature contains various ebl models with various merits ; a large fraction has been combined into different ebl scenarios by @xcite .
all of the scenarios that do not imply unreasonable intrinsic vhe spectra and most of the recent ebl models , produce similar gamma - ray opacities . here
, one particular ebl model is used as a reference , put forward by @xcite and referred to as the primack model in the following .
this model is compared in fig .
[ fig : ebl_models ] with ebl measurements and with the allowed ebl scenarios of @xcite .
although @xcite has proposed a number of models and the one chosen here is known to have slight problems above 60 @xmath2 m , it will be used for illustrative purposes only , with the exception of giving an estimate of the optical depth already present for low energy gamma rays in sect .
[ sec : upper_limits ] .
the attenuation of the intrinsic spectrum in the primack ebl model is shown in fig .
[ fig : primack_power_fit ] at the distances of the six blazars .
the figure shows the spectral slope , measured at 1 tev , becoming increasingly steep with redshift .
the steepening of blazar spectra with redshift was predicted by @xcite . in the case of pks 2155
, the absorbed spectrum could be fitted with an @xmath15 spectrum over the energy region from 0.3 tev to 3 tev , assuming an initial @xmath16 spectrum .
a linear fit to the spectral index steepened by the primack model , measured at 1 and 2 tev , vs. redshift , is shown in fig .
[ fig : primack_z_index ] . to determine the intercept of the linear fit , i.e. the intrinsic source spectral index , we assume that all blazars have the same intrinsic spectrum and adjust the level until the @xmath14 difference between measured and predicted indexes is minimized .
the spectral index at zero redshift predicted in this way is -1.62 and -2.2 , when the power law fits are performed at 1 tev and 2 tev , respectively .
considering the large degree of uncertainty in choosing an energy at which to measure the spectral index , fig .
[ fig : primack_power_fit ] implies that the intrinsic spectral index is on the order of -2 to -1.8 .
it should be noted that the detailed shape of the spectral steepening with increasing redshift depends strongly on the shape of the ebl .
for example , the seven possible ebl scenarios given in @xcite allow a range of slopes for spectral steepening with redshift , with the primack model roughly representing the average of the allowed ebl scenarios .
[ fig : primack_power_fit ] shows that in the primack ebl model , there is a leveling - off at high energies , not a simple power law with exponential cut - off as supposed above .
the cut - off energy in this case would correspond to the initial downturn at @xmath17 0.6 tev . at higher energies ,
the flux is so far diminished that measurements with vhe telescopes do not yield a statistically significant result .
note that if the optical depth is independent of energy , as is the case for @xmath18 , no change occurs at all in the spectrum with increasing redshift . in the primack model
this is almost the case between 2 tev and 5 tev , corresponding to the 2 - 5 @xmath2 m ebl .
consider a single vhe gamma ray of energy , @xmath19 , coming from a distant extragalactic source .
if the gamma ray makes a collision with another particle , it will not propagate in a straight line and is lost to the observer who is several mpc away .
pair production is the most likely type of inelastic collision in extragalactic space as the photon density , though varying with wavelength , is much higher than the matter density . for pair production to occur
, the total energy available must be greater than @xmath20 .
for vhe gamma rays with energy between 100 gev and 20 tev , the low energy photon must be in the range from 10 ev to 0.05 ev . the absorption of vhe gamma rays is described by a very simple rate equation @xmath21 . the number of vhe gamma rays , @xmath22 , with energy @xmath19 changes per unit time , @xmath23 , as @xmath24 where @xmath25 is the reaction rate defined below .
[ eq : population ] is solved by expressing @xmath23 in terms of the distance @xmath26 traveled by the photon : @xmath27 . for small redshift , @xmath28 , the relation @xmath29 holds so that the number of gamma rays at redshift @xmath30 is given by @xmath31 this defines the optical depth @xmath32 with @xmath33 km s@xmath8 mpc@xmath8 @xcite .
this formalism can been extended to cosmological distances @xcite .
the optical depth of the furthest detected vhe blazar h 1426 at @xmath34 , calculated according to eq .
[ eq : tau ] and with the cosmologically correct version differs by less than 10% . as this difference is less than the systematic uncertainty of vhe blazar spectra , eq .
[ eq : tau ] will be used . the measured flux , @xmath35 , is then given in terms of the emitted flux , @xmath36 , by @xmath37 , where @xmath38 is the measured energy .
this ignores the inverse - square law distance dependence ; it is of no concern as it effects only the absolute flux level and does not have an energy dependence .
the momentum distribution averaged pair production rate , @xmath25 , in units of [ @xmath39 , is @xmath40 where @xmath41 , @xmath42 is the angle between the incoming particles , @xmath43 is the azimuthal angle , and @xmath44 is the total energy .
the relative speed between the interacting particles is @xmath45 .
the comoving photon momentum distribution of the ebl is given by @xmath46 [ @xmath47 .
. [ eq : lambda ] can also be written in the more familiar way of particle density per energy per volume by substituting @xmath48 and with the replacement @xmath49 .
the spin - averaged pair production cross section , @xmath50 , is given by @xcite . from eq .
[ eq : lambda ] it can be seen that the optical depth is independent of the gamma - ray energy if the ebl photon density , @xmath51 , is independent of @xmath52 : i.e. @xmath53 only .
this corresponds to an energy density @xmath54 . on a @xmath55 plot ,
such an ebl spectrum would fall as @xmath56 and is quite possible in the optical / near - ir portion .
the sensitivity of the vhe gamma - ray spectrum to absorption by the ebl is best illustrated by the toy model of a monoenergetic ebl , i.e. isotropically distributed photons with only a single energy .
the absorption probability per unit length is given by @xmath57 where the constant @xmath51 is the co - moving photon density [ @xmath58 .
note that the only dependence on the ebl energy @xmath52 is through @xmath59 .
the absorption probability per unit length is plotted in fig .
[ fig : line_depth ] for various ebl energies ranging from ir to uv : the density is taken uniformly as 1 @xmath60 . though gamma - rays are absorbed most efficiently when @xmath61 ev@xmath62 , there is significant absorption of gamma rays with energies from half to four times as much . in general , knowledge of the optical depth
is not sufficient to unambiguously determine both the shape and magnitude of the ebl density .
if the ebl flux is known over a finite wavelength region , an infinite number of shapes are possible because of the limited energy resolution of this method .
the energy resolution is limited by the width of the pair production cross - section and the isotropic ebl photon distribution , see eq .
[ eq : lambda ] . if one assumes a shape for the ebl spectrum with only the overall normalization left as a free parameter , it is possible to determine the best fit ebl flux through a @xmath14 minimization between the measured and modeled optical depths .
however , this still relies on knowing the intrinsic blazar spectrum , which is not easily accomplished .
this method has been widely used to derive upper limits for power - law ebl spectra . to avoid specifying the ebl spectral shape altogether , upper limits on the true ebl density
are derived here by assuming the ebl to be monochromatic .
physically , we know that the ebl spectrum is extended . if we make that assumption that the ebl is monochromatic , the contribution of other wavelengths to the absorption of gamma rays is ignored .
hence , the ebl density at that single wavelength will have to be larger to reproduce the optical depth , than if other wavelengths were allowed to contribute to the absorption .
suppose we start with a known measurement of @xmath63 from a source at @xmath64 , illustrated in fig .
[ fig : tau_limit_tau ] ( _ left _ ) for three arbitrarily chosen values of @xmath63 .
then , we want to know : what ebl density does it take to reproduce this amount of absorption ?
the answer to this question depends on what photon wavelength is performing the absorption .
the least ebl density is required if it absorbs the gamma - ray most efficiently , that is , when @xmath61 ev@xmath62 .
the monochromatic ebl density corresponding to this case can be expressed in terms of the measured optical depth , @xmath65 , using eq .
[ eq : tau ] @xmath66 so that the upper limit ( ul ) on the true density is given by @xmath67 fig .
[ fig : tau_limit_tau ] ( _ middle _ ) shows the ebl flux corresponding to this case and fig . [ fig : tau_limit_tau ] ( _ right _ ) shows the corresponding optical depth produced by this single monochromatic ebl line .
consider then the case of many adjacent optical depth measurements , shown in fig .
[ fig : tau_limit_tau ] ( _ left _ ) at only three energies . as before ,
the monochromatic ebl flux is calculated in turn from each optical depth ( _ middle _ ) .
the total absorption , the sum of the three optical depths , is larger than was measured at the three energies ( _ right _ ) . hence , the monochromatic ebl flux is an overestimate of the true value required to achieve the measured attenuation .
determination of the amount of absorption present in one measured spectrum requires a priori knowledge of the intrinsic source spectrum .
though the mechanism for vhe gamma - ray production has been modeled , which model to choose is still unknown .
in addition , other important details are left open as well : what is the electron or proton spectrum ?
what is the strength of the magnetic field ?
what is the opacity in the vicinity of the source for gamma - rays to escape ? to place upper limits on the ebl density , the following simple assumption will be made : the intrinsic spectrum of all blazars are identical to a power law with index -1.8 ; this has been motivated in sect .
[ sec : comparison_of_spectra ] .
if the vhe spectrum has not been measured down to 0.2 tev , the flux at that energy is predicted from a power law fit extrapolated down from higher energies .
the small amount of ebl absorption present already at 0.2 tev is calculated from the primack model .
the assumed spectral index of -1.8 is very hard and can not extend indefinitely . by using a power law fit from higher energies , where the measured spectrum is steeper due to absorption , to extrapolate down in energy to 0.2 tev will likely overestimate the expected flux from the source .
the assumed intrinsic spectral index is approximately implied by the physics of tev blazars .
the intrinsic spectra of tev blazars are expected to peak in the multi - tev region @xcite ; hence the intrinsic spectral index should be somewhat harder than -2 .
detailed modeling show this to be the case for mrk 421 and mrk 501 @xcite .
@xcite show that after accounting for absorption by the ebl , the intrinsic spectra , in a @xmath55 representation , peak between 0.5 - 1.2 tev for mrk 421 , between 0.8 - 2.5 tev for mrk 501 , and between 1 - 5 tev for h 1426 . the expected flux in the absence of ebl absorption from each blazar is shown in fig . [ fig : spectra_agn_e2 ] by a dashed line .
the measured flux levels have been adjusted to the distance of mrk 421 using an inverse square law to show that the power output of the sources could be similar if the sources have identical geometries .
the optical depth is the ratio of expected flux to measured flux with the addition of the optical depth already present at 0.2 tev @xmath68 and this does not dependent on the overall normalization of the flux level .
the optical depth calculated in this way for the six blazars is shown in fig .
[ fig : tau_primack_data ] . also shown in the figure is the optical depth calculated in the primack model for comparison with the measured values .
a comparison of the measured optical depth for mrk 421 with the allowed ebl scenarios of @xcite is given in fig .
[ fig : tau_primack_dwek_mrk421 ] . here , the primack model lies roughly in the middle of the allowed ebl scenarios .
the measured optical depths of mrk 421 and mrk 501 are steeper in the 1 - 5 tev region than some of the ebl scenario and would favor a larger ebl density in the mid - ir region , as in the ir spectrum derived by @xcite .
the measured optical depths of the other sources are flatter , showing that this sample of six blazars is not completely homogeneous . the monochromatic ebl flux derived from each blazar spectral energy point under the above assumptions
is shown in fig .
[ fig : ebl_upper_limits ] . for comparison ,
the value of the monochromatic flux as calculated from the primack model is shown in the figure as well .
for the most part , the monochromatic flux values derived here agrees with the primack model ; this is also true for all of the other reasonable ebl scenarios .
differences in the derived monochromatic flux are likely caused by the intrinsic flare spectra of the sources being different from each other and not being pure power laws . because of the large distances to the blazars , it is highly unlikely that non - uniformities in the ebl give rise to the observed differences unless an ir galaxy is in the line of sight .
the limits derived from mrk 421 and mrk 501 are the lowest . to bring them higher and in line with the other sources ,
their intrinsic spectra would need to be slightly harder than -1.8 in the low energy region .
if the spectra were harder then curvature with a peak in the 1 - 2 tev range would make the flux limits more consistent with those from the other sources .
the systematic uncertainty in absolute energy calibration is 10% , directly corresponding to a 10% uncertainty in the ebl wavelength in fig .
[ fig : ebl_upper_limits ] . fig .
[ fig : ebl_limits ] shows the derived 98% confidence level upper limits together with optical ebl measurements and limits shown previously in fig .
[ fig : ebl_ir ] .
the range of upper limits from the six sources is indicative of the systematic error of this method .
the systematic error stemming from the optical depth at @xmath19 = 0.2 tev is not shown ; it is at the level of 5% at 1 tev and decreases with wavelength . the mid - ir limits on the ebl are monotonously decreasing with wavelength and above the limits inferred from fluctuation - analysis of the ebl by @xcite .
the upper limits in the optical to near - ir are in conflict with the detections claimed by @xcite and also with the hhh scenario by @xcite .
an increase in the upper limits in this region would be achieved if the intrinsic source spectrum peaked in the 0.7 - 2 tev region or if the optical depth at 0.2 @xmath2 m is substantially higher .
@xcite argues that his measurements are inconsistent with galaxy evolution models and much higher than what can be accounted for by the observation of galaxy populations . already at 2.2 @xmath2 m
the flux is higher than that claimed by @xcite from cobe and 2mass data .
they found that zodiacal emission modeling as their main uncertainty and responsible for the large fluxes .
the limits on the ebl derived here are compared with others , derived also with vhe gamma - ray observations , in fig .
[ fig : compare_upper_limits ] .
other upper limits are generally lower ; this is due to the assumption of a continuous ebl spectrum . in some cases ,
the upper limits are in conflict with measurements using optical methods .
the ebl is difficult to measure directly because of bright foreground radiation originating within the solar system .
vhe gamma - rays are very penetrating and shine through most of the dust in the galaxy and the solar system . on cosmological distances , pair production with optical and infrared radiation attenuates the gamma - ray flux .
therefore , if the vhe source spectrum is known , the ebl density can be inferred from the measured attenuation in the spectra .
unfortunately , the known extragalactic sources of vhe gamma rays are highly variable and modeling of the source spectrum is difficult . to estimate the intrinsic source spectrum , the strongest flare spectra of the six sources
were compared .
these were found to be similar once absorption by the ebl was accounted for .
therefore , the zeroth order assumption was made that the vhe flare spectra of bl lacs are actually the same ; with differences in attenuation arising solely due to the different distances to the objects . this simple , and almost experimental , way to infer the intrinsic spectrum at the chosen energy of 0.2 tev , results in a spectral index expected from the physics of tev blazars ,
see for example @xcite .
the method developed here to derive upper limits on the ebl does not hinge on an assumed ebl spectral shape . in detail
, it consists of the following three steps : ( 1 ) the optical depth for gamma - rays was derived at discrete spectral energy points from the difference between the measured gamma - ray spectrum and the assumed intrinsic spectrum . with better understanding of the intrinsic vhe blazar spectra , the optical depth , and hence the resulting upper limits , can be determined more accurately .
( 2 ) the density of ebl photon with a single wavelength was found that reproduces the measured optical depth at a given gamma - ray energy .
the ebl wavelength was fixed so that the pair production is most efficient , requiring the smallest density of ebl photons . ( 3 ) the derived monochromatic ebl density represents an upper limit to the true ebl density . the true ebl spectrum is extended and absorption of a gamma ray is produced by ebl photons over a range of wavelengths , thereby requiring a lower density than a monoenergetic ebl to produce the same optical depth .
the optical depths measured from the six blazars fall in line with predictions from a range of ebl models .
in particular , for mrk 421 the measured optical depths represent about the average of the allowed ebl scenarios given by @xcite .
the upper limits derived here are substantially above other limits that were derived assuming a continuous ( or stepwise continuous ) ebl , see fig .
[ fig : compare_upper_limits ] .
however , this does not mean that the limits derived here are weak ; to properly compare them with ebl models , upper limits must be calculated from the ebl models first .
this was done for one ebl model , shown in fig .
[ fig : ebl_models ] to represent an average of the allowed ebl scenarios , and the limits derived from the model and the measurements are in agreement .
the constraining power of the upper limits depends on the ebl spectrum : if the spectrum is steeply falling with wavelength , the upper limits are more conservative at shorter wavelength than if the ebl spectrum is flat .
the unambiguous and statistically correct determination of 98% ci upper limits from combining the results of the six blazars is difficult because an average intrinsic blazar spectrum was used , derived from the observed spectral steepening with redshift .
the upper limits derived here should be considered as a whole and not from one blazar alone .
the derived upper limits question one particular set of ebl measurements @xcite that are very high in the near infrared waveband and which galaxy formation models are typically not able to reproduce either . in principle
, one could extend the method described here to derive more constraining upper limits on the ebl density by reiterating a process of calculating upper limits and then using the upper - limit spectrum as an input to a second iteration of calculating upper limits .
to do better and measure the ebl density , requires accurate knowledge of the intrinsic and the absorbed source spectra over at least two decades of energy .
reduced error bars on the measured spectrum will be achieved with the new generation of telescopes coming online now .
however , the mechanism for production of vhe gamma rays , and hence the intrinsic source spectrum , is still under considerable debate .
unless a theoretical model is accepted or a `` standard candle '' at cosmological distances is identified with a well understood gamma - ray spectrum , the spectra from all these sources will only provide upper limits on the ebl flux . only through further observation at all wavelengths
will the gamma - ray production mechanism be understood .
the glast satellite , to be launched in 2006 , will cover the energy range from 0.1 - 100 gev .
together with the low threshold of about 100 gev for the new arrays of imaging cherenkov telescope such as hess , veritas , and cangaroo , simultaneous flux measurements can be performed over almost six decades in energy .
the author acknowledges the award of a smithsonian predoctoral fellowship and the support of the u. s. department of energy .
also , i thank t. c. weekes for helpful discussions and the anonymous referee for valuable comments . | the direct measurement of the extragalactic background light ( ebl ) is difficult at optical to infrared wavelengths because of the strong foreground radiation originating in the solar system .
very high energy ( vhe , e@xmath0100 gev ) gamma rays interact with ebl photons of these wavelengths through pair production . in this work ,
the available vhe spectra from six blazars are used to place upper limits on the ebl .
these blazars have been detected over a range of redshifts and a steepening of the spectral index is observed with increasing source distance .
this can be interpreted as absorption by the ebl . in general , knowledge of the intrinsic source spectrum
is necessary to determine the density of the intervening ebl .
motivated by the observed spectral steepening with redshift , upper limits on the ebl are derived by assuming that the intrinsic spectra of the six blazars are @xmath1 .
upper limits are then placed on the ebl flux at discrete energies without assuming a specific spectral shape for the ebl .
this is an advantage over other methods since the ebl spectrum is uncertain . |
the study of rare gas and metal clusters interacting with intense infrared , optical , and ultraviolet laser pulses has emerged as a new promising research area in strong field physics ( see @xcite for reviews ) .
the generation of fast electrons and ions , the production of high charge states , the generation of x - rays and nuclear fusion ( in the case of deuterium - enriched clusters ) was observed using cluster targets in strong laser fields of intensities up to @xmath0w/@xmath1 .
hot and dense nonuniform , nonequilibrium and nonstationary plasmas produced under such conditions and confined on a femto- or picosecond time scale to nanometer sizes so - called _
nanoplasmas_are new physical objects with unusual properties .
one of the most important features of nanoplasmas is the very efficient energy transfer from light to charged particles , which is much higher ( per particle ) than for an atomic gas of the same average density @xcite . although the particular mechanisms responsible for the laser energy deposition in clusters still remain debated @xcite , the pivotal role of collective plasma dynamics ( in particular the excitation of surface plasmons ) and of nonlinear resonance ( a definition is given below in sec .
[ reoloh ] ) in the energy transfer from laser light to the electrons in the nanoplasma and the subsequent outer ionization was proved in experiments , simulations , and simple analytical models @xcite .
while the energy absorption from intense laser pulses by nanoplasmas has been widely studied both in experiments and theory , much less attention has been paid to harmonic emission from such systems .
naturally the question arises whether laser - driven clusters , being highly nonlinear systems , may be a source of high - order harmonics as efficient as atoms in the gas phase ( or even more efficient ? ) .
there are at least two essentially different physical mechanisms which could be responsible for emission of harmonics from such a system .
first , the mechanism based on recombination of the virtually ionized electron with its parent ion @xcite , well - studied in gaseous atomic targets , may also work in clusters where it can be modified by the fact that the atoms are closer to each other , so that the electron s motion between ionization and recombination can be distorted by the field of the other ions .
moreover , the electron may recombine with an ion different from its parent .
this leads to modifications in the single - electron dynamics and in the phase matching conditions while the physical origin of harmonic generation ( hg ) remains the same as in a gas jet .
modifications of the atomic recombination mechanism in clusters have been considered in @xcite .
however , in high intensity fields where each atom is loosing one or several electrons already during the leading edge of the laser pulse the recombination mechanism can hardly be efficient .
second , as dense electron plasmas is produced inside a cluster , its coherent motion may be an efficient source of radiation . in a collisionless plasma , as it is generated by intense laser fields inside small clusters , individual electron - electron and electron - ion collisions can not destroy the coherency of the collective electron motion .
this coherent , collective electron motion may cause hg if it is nonlinear . recently , several experiments on high harmonic emission from plasma surfaces illuminated by short laser pulses of intensity @xmath2w/@xmath1 and higher were reported ( see , e.g. , @xcite ) . the physical picture which is behind such _ plasma harmonics _ appears to be more complicated and diverse than the recombination mechanism in atomic hg . in macroplasmas
the magnetic component of the lorentz force is a typical source of nonlinearity . in this case
the nonlinearity parameter is @xmath3 where @xmath4 is the typical velocity related to collective oscillations of the electron plasma and @xmath5 is the speed of light . as a consequence , generation of harmonics from dense macroplsmas may require relativistic intensities @xcite .
as compared to macroplasmas , clusters introduce an extra source of nonlinearity due to their small spacial size , namely @xmath6 , where @xmath7 is the cluster radius and @xmath8 is the amplitude with which the electron cloud oscillates under the action of the laser field @xcite .
therefore , one could expect strongly nonlinear electron motion even in the nonrelativistic regime .
it should be noted , however , that although individual collisions may not be important , there are other effects which can spoil the coherency required for efficient radiation . most of these undesirable effects can be attributed to dynamical instabilities induced by the interaction of particles with the mean self - consistent field ( in the presence of the laser field ) .
therefore , the examination of _ plasma harmonics _ usually requires not only the analysis of the collective motion but that of individual electron trajectories as well . up to now only a few experiments on hg from clusters
are known . in refs .
@xcite hg from rare - gas clusters irradiated by infrared pulses of moderate ( @xmath9w/@xmath1 ) intensity was studied .
it was shown that under such conditions harmonics can be generated up to higher orders and with a higher saturation intensity than in a gas jet .
in addition other interesting properties including different power laws in the intensity - dependent harmonic yield for particular harmonics for clusters and atoms were reported @xcite . however , in refs .
@xcite the applied intensities were rather low to create a dense nanoplasma inside clusters and the main features of the recorded spectra ( like a plateau followed by a cutoff ) were found to be quite similar to the case of a gaseous target .
this allows to attribute the observed effects to hg along the standard atomic recombination mechanism as described above .
recently the experimental observation of the third harmonic ( th ) generation from argon clusters in a strong laser field was reported @xcite .
the laser intensity was varied between @xmath10 and @xmath11w/@xmath1 and was thus sufficiently high for the creation of a dense nanoplasma . at such intensities hg along the atomic recombination mechanism relevant in gases
is essentially suppressed because of saturated single - electron ionization so that the observed th signal can be fully attributed to the nonlinearity of the laser - driven nanoplasma . a resonant enhancement of the th yield ( when the mie frequency of the expanding cluster approaches three times the laser frequency ) has been measured using a pump - probe setup
. the th enhancement of the single - cluster response , as studied in theory before @xcite , is , however , masked in the experiment by phase matching effects whose optimization at high average atomic density necessary to create clusters has been shown to be more intricate than for rare gas jets .
the latter complication makes an experimental study of nanoplasma radiation a difficult task while in computations it can be simplified by first examining the single - cluster response and , second , analyzing propagation effects . in the recent paper @xcite , we considered the radiation emitted by a single cluster exposed to a strong laser pulse .
we computed harmonic spectra from argon clusters in short 800-nm pulses of intensity @xmath12w/@xmath1 using a 3d particle - in - cell ( pic ) code ( applied before to the study of collisionless energy absorption in laser - driven nanoplasmas @xcite ) .
the most intriguing outcome of our study was the absence of high - order harmonics in the computed spectra , even at high intensities for all cluster sizes we considered .
we attributed this effect to the above - mentioned dynamical instability in motion of individual electrons . in this paper
we introduce two analytical models which describe the collective and single - electron dynamics of a laser - driven nanoplasma , respectively , and apply them for the explanation of both the numerical results @xcite and the data @xcite .
we show that the numerical results can be well described and understood within the rigid sphere model ( rsm ) but only for low harmonic orders @xmath13 while for higher harmonics the rsm yields qualitatively wrong predictions . using a simple 1d model we describe a stochastic , resonant single - particle electron dynamics which suppresses the emission of high harmonics .
the paper is organized as follows . in sec .
[ softpapr ] we formulate the statement of the problem and describe the numerical method and the pic results . in sec .
[ reoloh ] the spectra extracted from the pic simulations are compared with the predictions of the rsm . in sec .
[ sohhvsboie ] we introduce a model for the description of the single - electron dynamics and radiation and use it to explain the suppression of the high harmonic yield .
the last section contains the conclusions .
a cluster is converted into a dense electron plasma almost promptly if the laser pulse is intense enough to ionize the cluster constituents .
we refer to this process as inner ionization which should not be confused with outer ionization , when the nanoplasma electrons leave the cluster .
several competitive processes govern the evolution of this plasma during the interaction with the pulse and later , until the cluster becomes dissolved due to coulomb explosion or hydrodynamic expansion . the electron density increases because of further ionization of atoms and ions by the local electric field which may differ essentially from the applied laser field . with increasing plasma density due to inner ionization ,
the oscillating electric field of an infrared laser is screened , so that its amplitude inside the cluster may be a few times or even an order of magnitude less than the amplitude of the incident wave ( see more explanations on screening in sec .
[ sohhvsboie ] below eq.([e0 ] ) ) . simultaneously , as soon as a sizeable fraction of electrons have left the cluster , a quasistatic space charge field is built up which may become strong enough to induce further inner ionization ( `` ionization ignition '' , @xcite ) . on the other hand , both outer ionization and the expansion of the cluster reduce the electron density .
the net result of this competition is very sensitive to all parameters , including laser intensity , pulse duration , cluster size and type of atoms .
however , for the vast majority of parameters a significant part of the electrons remains confined within the expanding ionic core . during this stage of the cluster evolution until the laser pulse is off the nonlinear motion of the laser - driven nanoplasma may cause emission of laser harmonics .
we have observed the above described scenario in pic simulations of laser - cluster interaction , as reported in @xcite .
the dynamics and the radiation of nanoplasmas was studied for @xmath14 clusters ( with the number of atoms @xmath15@xmath16 and radii @xmath17@xmath18 nm ) , irradiated by linearly polarized , 8-cycle sin@xmath19-laser pulses with an electric field @xmath20 and the wavelength @xmath21 nm . here
@xmath22 is the carrier frequency and @xmath23 is the number of optical cycles in the pulse .
the contribution of electrons remaining bound in atoms ( ions ) during the interaction was not taken into account .
results were reported in ref .
@xcite where also more details about the numerical simulation may be found .
here we restrict ourselves to briefly summarize the main results : * the relative yields of low order harmonics depend on the laser intensity and the cluster size . *
the time - frequency analysis of harmonic spectra shows that low order harmonic enhancements occur when multiples of the laser frequency resonate with the transient mie frequency . * even for the very high intensity @xmath24w/@xmath1 no distinguishable high harmonics ( higher than the 7th ) appear in the spectra ( see fig . 1 in ref .
@xcite ) . * only a part of the nanoplasma deeply bound inside the ion core contributes to ( low order ) harmonic generation .
the enhancements of particular low - order harmonics predicted in @xcite have been later studied numerically for small clusters using molecular dynamics simulations @xcite and were finally observed in the experiment for the case of the third harmonic @xcite .
it is well established now that the physical origin of enhancements is the resonance between the harmonic frequency and the mie frequency of the expanding nanoplasma .
the time - frequency analysis of the radiation as calculated from the pic results for the total acceleration further confirmed this statement @xcite .
typical tf diagrams , showing which frequencies are emitted when , are reproduced in fig .
[ fig1 ] . in the same plots we show the scaled time - dependent effective mie frequency @xmath25 .
the standard definition of the mie frequency @xmath26 , where @xmath27 is the average ion charge and @xmath28 is the atom density , is only appropriate for the case of an almost homogeneous charge distribution ( in this section atomic units @xmath29 are used ) .
however , because of the ignition effect and the cluster expansion the average ion charge @xmath27 depends on the ion position . as we found in @xcite , the low - order harmonics are emitted mainly by electrons confined within the ion core and with excursion amplitudes comparable to the initial cluster radius @xmath7 or less . in the following
we will refer to these electrons as the _ deeply bound _
electrons , and we define the effective mie frequency as @xmath30 with @xmath31 the total ionic charge inside the sphere of radius @xmath7 within which the cloud of deeply bound electrons oscillates
. cluster at laser intensities @xmath32w/@xmath1 ( a ) , @xmath24w/@xmath1 ( b ) .
panel ( c ) shows the tf diagram for an @xmath33 cluster at the intensity @xmath24w/@xmath1 .
the scaled time - dependent mie frequency @xmath25 is included in all plots ( solid lines ) .
[ fig1],scaledwidth=40.0% ] it is clearly seen from the tf diagrams of fig.[fig1 ] that the enhancements of the third and fifth harmonic must be attributed to the respective high - order nonlinear resonances between the laser frequency and the mie frequency , @xmath34 ( with @xmath35 ) . here
we call the resonance _ nonlinear _ since it appears in a _ nonlinear _ system .
this nonlinearity ( i.e. , anharmonicity in the effective potential of electron - ion interaction ) is a necessary condition for the emission of harmonics .
the physical origin of this nonlinearity may be either the electrons jutting out of the core during their motion , thus sensing the coulomb tail @xcite , or the inhomogeneous charge distribution within the ion core @xcite .
note that the appearance of a significant signal at frequencies different from odd multiples of the laser frequency ( namely , the second harmonic in fig.[fig1](a ) and the forth harmonic in fig.[fig1](c ) ) should be attributed to the excitation of eigenoscillations of the electron cloud with the time - dependent eigenfrequency @xmath36 .
these oscillations are of significance only in short pulses , as used in the pic simulations . the results of the tf analysis in ref .
@xcite not only explain , at least qualitatively , the resonant enhancement of the th observed in @xcite but also confirm the idea formulated there that with a proper adjustment of parameters harmonics of higher order ( 5th , maybe 7th ) could also be resonantly enhanced .
it was shown in earlier studies @xcite that the resonant enhancements of low - order harmonics may be reasonably described using a very simple rigid sphere model ( rsm ) . here
we show that , while for low - order harmonics the rsm works well , it fails even qualitatively to reproduce the high - energy part of the spectrum . in a rsm
@xcite it is usually assumed that both ions and electrons form homogeneous , rigid spheres with sharp boundaries . in this case
the electron and ion charge density distributions are _
e(i)(r)=n_0(r_0-r ) , [ rho0 ] where @xmath28 is the atom density in the cluster and @xmath37 is the heaviside step function .
the signs @xmath38 correspond to electrons and ions , respectively . within this model
the restoring force @xmath39 depends upon the displacement @xmath40 of the electron cloud as @xcite @xmath41 g_0(y)=()^2\ { ly-+ , 0y2 , + , y2 . .
[ g0 ] here the dimensionless coordinate @xmath42 and force @xmath43 are introduced .
the equation of motion reads = * f*(*y*)-- [ ne0 ] where @xmath44 is the electric field of the laser wave and @xmath45 is the effective damping constant which can be estimated assuming a collisionless damping mechanism @xcite ( in calculations we use @xmath46 , in accordance with such estimates ) . using the rsm significant insights into the absorption mechanisms in clusters @xcite and th emission @xcite
have been gained . however , an essential shortcoming of the rsm in the form specified by eqs.([rho0]),([g0 ] ) is the appearance of even powers of @xmath47 in ( [ g0 ] ) at @xmath48 .
physically it is the result of the discontinuous charge distributions ( [ rho0 ] ) at @xmath49 . as a consequence , in a perturbative treatment of harmonic generation the intensity of the @xmath50-th
harmonic is proportional to @xmath51 instead of @xmath52 .
for example , for the third harmonic one obtains with the rsm @xmath53 , which is obviously unphysical @xcite . to improve the model it is sufficient to assume a smoothed charge distribution for the electrons while the positive charge density still may be described by ( [ rho0 ] ) .
we use a gaussian of characteristic width @xmath54 for the electron charge density , _
e(r)=-n_0(-r^2/r_e^2 ) , [ rho1 ] and assume that the net charge density is zero at the cluster center
. from ( [ rho1 ] ) the total number of electrons is @xmath55 .
the restriction @xmath56 gives @xmath57 .
the equation of motion has now the same form ( [ ne0 ] ) but with a modified restoring force : @xmath58 @xmath59-(y^3 - 1){\rm erf}[a(y-1)]\bigg\rbrack+\right . \nonumber\ ] ] @xmath60-\right . \nonumber\ ] ] .-e^-a^2(y+1)^2[1 - 2a^2(1-y+y^2 ) ] } [ g1 ] where @xmath61 and @xmath62 is the error function .
contrary to eq .
( [ g0 ] ) , a decomposition of the restoring force ( [ g1 ] ) contains only odd powers of @xmath47 while it has the same asymptotic coulomb behavior for large displacements .
the respective asymptotic expansions have the form [ compare with ( [ g0 ] ) ] g_1(y)=()^2\ { la_1y+a_3y^3 + ... , y1 , + + o(e^-a^2y^2 ) , y1 .
[ g1as ] with @xmath63 and @xmath64 .
it is clear that @xmath65 if @xmath66 , i.e. , when the tail of the electron density distribution does not stick out of the ion core . in this case
the eigenfrequency of small oscillations @xmath67 is equal to @xmath68 . for a finite @xmath69
the spread of the electron cloud beyond the ion core reduces the eigenfrequency .
note that within the model we describe above the restoring force is nonlinear because a part of the electron cloud spreads out of the ion core . in this case
the leading term in the nonlinear part of the force can be estimated as @xmath70 . as a result ,
the third harmonic signal scales as @xmath71 with the cluster radius , in contrast to the standard @xmath72 low expected for a volume bulk effect .
this shows that the above discussed low - order harmonic emission from the nanoplasma is a _ surface _ effect which becomes relatively less pronounced with increasing of the cluster size . in small cold metal clusters subject to infrared laser pulses of moderate intensity
the @xmath71 dependence was observed in @xcite though the experiment was not entirely conclusive ( see also @xcite for the respective theoretical analysis ) .
figure [ fig2 ] shows harmonic spectra @xmath73 calculated from the above - described rsm . within the model , the
dynamics and radiation of the electron cloud are governed mainly by the values of the two parameters @xmath74 and @xmath75 , determining the possibility of resonant enhancements for particular harmonics and the number of excited harmonics in the spectrum .
the eigenfrequency of the electron cloud @xmath76 depends upon the amplitude of the oscillations ( the maximum value @xmath77=\sqrt{a_1}\omega_{\rm mie}$ ] corresponds to small harmonic oscillations ) and , therefore , upon the laser intensity . as a result , at fixed cluster parameters the resonant enhancements appear at certain values of the intensity . for the spectra of fig.[fig2](a ,
d ) and ( c , f ) the parameters were chosen such that the maximum value of the eigenfrequency exceeds the integers 3 and 5 , respectively , so that , with increasing intensity resonant enhancements of the third and the fifth harmonic appear .
these enhancements are clearly seen from the comparison of these spectra with the ones calculated for the case shown in fig.[fig2](b , e ) when the maximum eigenfrequency @xmath77\approx 4.4 $ ] is far from both resonances . in panels
( a - c ) the intensities are chosen such that the sphere s oscillation amplitudes are almost the same in each case so that the relative difference in the strength of the third and the fifth harmonic ( about three orders in magnitude ) shows the efficiency of the resonant enhancement .
for the higher intensities in panels ( d - f ) these enhancements are even more pronounced .
a comparison between the rsm and the pic results shows that for the low - order harmonics the rsm provides a qualitatively correct description of the spectrum , including the effect of resonant enhancement .
this gives an additional justification of the rsm for applications to laser - driven nanoplasmas .
however , for relatively high harmonics ( ninth or higher ) the results predicted by the rsm appear to be qualitatively wrong . indeed , the rsm predicts the appearance of higher harmonics with increasing laser intensity , as it is seen from fig .
[ fig2](f ) where the spectrum ends up at the 13-th harmonic . by a moderate variation of the parameters inherent to the model , namely @xmath78 and @xmath69
, one may obtain even higher harmonics for the same intensities used for the spectra of fig .
as expected for a classical system , no signature of a plateau emerges in the spectra .
however , the pic simulations reported in @xcite do not show any harmonics above the seventh within the studied domain of parameters , including the very high intensity of @xmath24w/@xmath1 .
the same effect was found in @xcite where no harmonics above the ninth have been observed for intensities up to @xmath2w/@xmath1 . in ref.@xcite
we concluded from the inspection of individual pic electron s trajectories that a dynamical instability induced by the resonant interaction of electrons with the time - dependent self - consistent field is responsible for the suppression of high harmonics .
obviously , such a mechanism is beyond the rsm since the latter accounts for collective electron dynamics only . in the next section
we introduce a single - particle model suited for the analysis of the motion and the radiation of individual electrons so that the instability responsible for the suppression of high harmonics from cluster nanoplasmas under the above - described conditions can be studied .
it is known from numerical studies ( see examples in @xcite ) that in a laser - driven cluster the electron population separates into a dense core with a radius comparable to the initial cluster radius @xmath7 and a rarefied halo with a typical size of several @xmath7 .
this subdivision is equivalent to a separation of quasifree electrons into deeply and weakly bound electrons , correspondingly .
a decrease of the electron density due to outer ionization is compensated by the inner ionization of atoms , provided inner ionization is not yet depleted , so that the cycle - averaged electron density distributions both in the core and the halo evolve rather slowly in time .
the core oscillates with relatively small deformations so that the rsm seems to be applicable to the deeply bound electron s dynamics while electron trajectories in the halo are strongly disturbed by the laser field and can not be captured by the rsm . in ref .
@xcite the radiation of individual pic electrons has been considered ( see fig . 3 there ) .
it was shown that electrons radiate harmonics as long as they move inside the dense core .
being liberated from the core , electrons leave the cluster vicinity almost promptly , usually within a laser period .
this means , that the halo contains basically no permanent population , but consists almost entirely of electrons on their way out of the cluster .
hence the electron density in the halo determines the rate of outer ionization from the cluster . during the ejection
, each electron emits an intense flash of radiation with an almost continuous spectrum that extends up to significantly higher frequencies than present in the net harmonic spectrum . in ref.@xcite
we argued that these flashes add up incoherently in the total emission amplitude .
here we introduce a simple analytical 1d model which helps to illuminate the physical origin of this incoherency . despite its simplicity our model is able to describe , at least qualitatively , all essential features seen in the simulations .
let us suppose frozen ions and the number of nanoplasma electrons fixed so that in the absence of the laser field each electron moves in the time - independent self - consistent potential @xmath79 .
the laser field excites oscillations of the electron cloud which induce the ac part of the space charge field @xmath80 .
the net oscillating field inside the system can be written as ( t)=e_l(t)+e^(i)_sc(t)_0f(t)(_l t+ ) , [ field ] where @xmath81 is the time - dependent pulse envelope . if the mie frequency notably exceeds the laser frequency , @xmath82 , as it is the case for infrared lasers , the laser field and the ac space - charge field @xmath80 essentially compensate each other . in this case
the amplitude of the net oscillating field inside the cluster @xmath83 is related to the amplitude of the laser field @xmath84 according _ 0e_0 .
[ e0 ] this result is commonly referred to as _ screening _ of a low - frequency laser field inside clusters ( see , e.g. , the review @xcite ) .
there is no contradiction between this screening of the laser field for individual electrons and the fact that the whole electron cloud feels the unscreened field . indeed , in the rsm there are two forces acting on the electron cloud : one due to the interaction with the ion core , another one due to the laser force , see eq.([ne0 ] ) . in the single - electron description
we should also take into account the interaction between the electron under consideration and all other electrons in the cloud . within the model
we assume that the electron cloud undergoes small , slightly nonlinear oscillations , so that this extra force is almost homogeneous inside the cluster and oscillates in time with the frequency @xmath22 . within the rsm and under the condition @xmath85 we assume throughout the paper that the electron cloud displacement @xmath86 reads @xmath87 calculating the electric field induced inside the cluster due to this displacement and summing it up with the laser field ( [ field ] ) one obtains the estimate ( [ e0 ] ) for the amplitude of the net oscillating field . this type of screening results from the coherent superposition of the applied and the self - consistent field and has nothing in common with damping of electromagnetic waves in macroplasmas .
the latter occurrs on the spatial scale of the skin - depth , which is in general much bigger than the typical cluster size we consider .
taking a cluster consisting of @xmath88 ar atoms ( @xmath89 nm ) with the average ion charge @xmath90 and the degree of outer ionization @xmath91 , one can estimate @xmath92ev .
in a ti : sa laser pulse of intensity @xmath93w/@xmath1 ( @xmath94a.u . )
the amplitude of the oscillating field inside the cluster is according to ( [ e0 ] ) @xmath95a.u .
, i.e. , more than one order of magnitude below the amplitude of the applied laser field . the quasistatic part of the space - charge field @xmath96 which traps electrons within the ion core can also be estimated for the assumed values of @xmath97 and @xmath98 .
namely , the field near the cluster edge is @xmath99a.u . , i.e. , more than one order of magnitude above the oscillating field amplitude . from this estimate
we conclude that the oscillating field inside the cluster usually remains small compared to the quasistatic space - charge field . within the model the electron s evolution
is governed by the hamiltonian h(p , x , t)=+u(x)-ee(t)xh_0(p , x)-ee(t)x [ h ] and the corresponding newton equation = mx=-+ee(t)-ee^(ii)_sc(x)+ee(t ) , [ ne ] where @xmath100 and @xmath101 are the electron mass and charge .
the well @xmath79 is created by the quasistatic part of the space charge .
we model it by the function u(x)=u_0[1 - 1/ ] , [ c ] where the values @xmath7 and @xmath102 are the cluster radius and the depth of the self - consistent well , respectively . here
we choose the energy minimum @xmath103 , so the @xmath104 is the continuum threshold . for small excursion amplitudes
this well is a nonlinear oscillator while for large excursions it has the desired coulomb behavior .
according to the estimates given above we assume the inequality 1 , f_0= [ mu ] being satisfied , where @xmath105 has the meaning of a characteristic quasistatic force trapping the electron .
the dynamics of the unperturbed system with the hamiltonian @xmath106 is characterized by the energy dependence of the eigenfrequency ( ) = , t()=_a^b , [ t ] where @xmath107 is the oscillation period , @xmath108 , @xmath109 are the turning points , and @xmath110 is the total energy @xcite .
the energy - dependent parameter ( ) = [ beta ] characterizes the nonlinearity of the unperturbed system and thus its potential capability to emit harmonics .
figure [ fig3 ] shows the energy dependence of the scaled eigenfrequency @xmath111 and the parameter ( [ beta ] ) . in cluster potentials the period @xmath107 increases with increasing energy so that d / d0 , b= [ hfinal ] with @xmath112 .
the electron behavior can thus be qualitatively described as nonlinear oscillations in @xmath113 space , known as _ phase oscillations _ @xcite . in the new canonical variables which have the formal status of momentum @xmath114 and coordinate @xmath115 the phase space of the hamiltonian ( [ hfinal ] ) splits into domains of finite and infinite motion ( see fig.[fig4 ] ) .
finite motion corresponds to a particle trapped by the resonance , while particles moving infinitely do not intersect with the resonance .
the separatrix of ( [ hfinal ] ) is a boundary between these two domains .
the motion near separatrices is unstable , so that even a small variation in the initial conditions may entirely change a trajectory . as a result
, particles approaching a separatrix may penetrate from one domain to another or may be trapped by a resonance .
the parameters characterizing the motion of trapped particles are the maximum deviation from the resonant action @xmath116 and the frequency of small phase oscillations : p_max=2 , _ ph=_l .
[ omph ] in energy space the positions of the separatrices of the first - order resonance are determined by _ 1^=(i_1p_max)=_1_1 , _ 1u_0 .
[ threshold ] if the energy intersects a respective threshold so that @xmath117 , the electron becomes trapped by this resonance domain and experiences phase oscillations with the frequency and amplitude both proportional to @xmath118 . due to the appearance of a new time scale given by the frequency of the phase oscillations ( [ omph ] )
the motion becomes aperiodic and highly nonlinear .
it should be emphasized that in weak fields the perturbation parameter @xmath119 is far from resonance and @xmath118 close to it so that the near - resonant motion appears to be much more perturbed than the off - resonant one .
the interaction with an isolated resonance can not lead to ionization since the electron remains trapped for an , in principle , infinite time @xcite .
however , the higher - order resonances lying above may come into play .
as soon as the separatrices of neighboring resonances intersect , the particle , captured by the first resonance , may jump to the third , etc .
because the volume of the accessible phase space is increasing with energy this inter - resonance motion will have a predominant direction , namely towards higher energies .
this leads to a fast liberation of the electron from the system , known as _
stochastic ionization _ @xcite . for realistic parameters of laser - cluster interaction
this overlap of resonances is realized with almost 100% probability so that cases where the particle remains trapped by a resonance are rare while almost prompt ionization occurs as soon as the first - order resonance is reached . we visualize the above - described scenario by solving eq.([ne ] ) numerically . in the calculation
we take @xmath120 and @xmath121 so that even the amplitude @xmath122 ( @xmath123 ) still corresponds to the weak - field regime , as defined above .
these parameters are not arbitrarily chosen .
indeed , a solution of eq.([ne ] ) with the well ( [ c ] ) and the field ( [ field ] ) with a slowly varying envelope depends on four dimensionless parameters , @xmath119 , @xmath124 , @xmath125 and @xmath126 , the two last of them defining the initial conditions . to recalculate all the parameters for a real system
we should assume some certain values for the cluster radius and the laser field frequency , which then define all other parameters .
for typical values , say @xmath127 nm and @xmath128ev , and for the dimensionless parameters of fig .
[ fig5 ] one may check that the resulting strengths of the quasistatic and the oscillating parts of the self - consistent field indeed correspond to the estimates given below eq.([e0 ] ) .
the positions of the most important first and third - order resonances are @xmath129 and @xmath130 , respectively .
a particle with the initial energy @xmath131 starts its motion at @xmath132 and @xmath133 when the field @xmath134 is negligibly small .
then the field ( with a gaussian envelope ) increases , and the electron propagation under the action of the full force is calculated until @xmath135 . by choosing different phases @xmath126 of the field we model different initial conditions for the particle at the fixed initial energy @xmath136 .
the results are summarized in fig.[fig5 ] .
figure [ fig5]a corresponds to the `` perturbative '' regime of interaction .
the initial energy is far enough from the first ( lower ) resonance , so that the trajectory in the energy space does not intersect the respective lower separatrix , or just touches it . as a result ,
the trajectory remains weakly disturbed , its shape is well described as a superposition of oscillations with the frequencies @xmath137 and @xmath22 . by choosing different initial conditions we obtain trajectories simply shifted in time by the value of @xmath126 .
figures [ fig5](b , c ) correspond to the `` resonant '' regime of interaction where the initial energy is high enough or the field is strong enough to cause penetration of the particle into the vicinity of the first resonance .
the five trajectories plotted in figs.[fig5](b , c ) show that the near - resonant motion is very sensitive both to the initial conditions and to the field amplitude so that a particular trajectory appears to be unpredictable .
usually the particle is emitted from the system while in rare cases it remains in a bound state after the pulse is off , being trapped by a resonance ( see the trajectories in fig .
[ fig5](b , c ) ) . from this observations
we may conclude that at parameters typical for intense laser - nanoplasma interactions the particle behavior in the `` resonant '' regime becomes stochastic , as it is expected to be according to the general theory @xcite . in classical ionization , a particle has to overcome the potential barrier , i.e. , its total energy must exceed the maximum value of the potential energy suppressed by the field at some time instant .
obviously , the trajectories of fig .
[ fig5](b , c ) satisfy this condition , while both trajectories of fig .
[ fig5]a do not . in fig .
[ fig6 ] we show the trajectories in energy space evaluated for half the laser frequency , @xmath138 , and all other parameters the same as in fig .
[ fig5]b .
the plots show that with decreasing laser frequency the time - dependent energy gain from the field to the particle decreases down to a perturbative level , the total energy remains always below the maximum of the time - dependent potential energy , and no ionization or excitation occurs . within the resonant picture
we exploit here the qualitative difference between the trajectories of figs .
[ fig5](b , c ) and [ fig6 ] appears because with decreasing frequency the first resonant level is shifted up and no penetration into the resonant area between the separatrices takes place anymore .
it is instructive to show the connection between the single - electron model specified by eqs .
( [ field]),([h]),([ne]),([c ] ) and a model which describes collective motion , as the rsm of sec .
[ reoloh ] does .
the rsm deals with the electron cloud displacement @xmath86 whose fourier - transform is directly related to the spectrum .
this value can also be calculated within a single - electron picture as * x*(t)=d_0d * r*(_0,,t)f(_0 , ) , [ xt ] where @xmath139 is the individual trajectory with the initial energy @xmath136 and the initial condition @xmath126 , and @xmath140 is the distribution function for electrons before the field is on .
one should note that , contrary to the rsm , the spatial distribution of the electrons in the presence of the field is not known unless one calculates all individual trajectories . calculating a trajectory
@xmath139 analytically is possible only within perturbation theory with respect to the external field where one may easily derive ( [ ne0 ] ) with the linear part of the restoring force only from ( [ xt ] ) .
the derivation of nonlinear corrections , although doable , requires very cumbersome algebra .
the single - particle model is appropriate for a qualitative description of the stochastic resonant behavior but hardly applicable to the study of the slightly unharmonic motion of the deeply bound electrons .
the analysis of the previous subsection gives a direct explanation of the radiation spectra extracted from the pic results in ref .
@xcite both for individual electrons and for whole clusters .
deeply bound electrons with energies below the first resonant separatrix @xmath141 move along slightly perturbed regular trajectories .
this causes hg with rapidly decreasing yield as a function of the harmonic order so that even the seventh harmonic is barely present in the corresponding spectrum of fig .
an individual electron , while passing the resonance and being trapped by it or leaving the cluster potential , emits radiation due to its strong acceleration , seen as a flash in the tf spectrograms of fig .
[ fig7]b , c .
these spectrograms should be compared with the ones extracted from our pic results ( see fig.3 in @xcite ) .
exactly because of the stochastic nature of nonlinear resonance the electrons trajectories are very sensitive to the initial conditions with which the nonlinear resonance is entered , as is clearly seen from fig .
[ fig5]c where solid black , dashed black and gray trajectories correspond to the same initial energy and the same field amplitude but different phases of the external field . as a result , flashes from different electrons are incoherent ( the corresponding amplitudes have nearly random phases ) , and , being added up , vanish in the total dipole acceleration .
a and both curves of fig .
[ fig5]b ) electron trajectories.,scaledwidth=40.0% ] this shows that exactly the same mechanism behind efficient energy absorption by and outer ionization from clusters , namely nonlinear resonance @xcite , restricts hg from them by breaking the coherent electron motion once it becomes strongly anharmonic .
only well - bound electrons trapped inside the ionic core with energies far from the resonance contribute to the net , coherent radiation of the cluster .
a similar behavior was observed in classical ensemble simulations of atomic hg @xcite .
instabilities induced by the resonant interaction grow in time ( usually with an exponential rate ) .
thus the picture depends also on the pulse duration . with all other parameters fixed , for longer pulses , more and more trajectories from the vicinity of the separatrix experience a stochastic behavior . as a consequence ,
an increase of the pulse duration should lead , in general , to a further loss of coherency .
although laser - irradiated cluster nanoplasmas emit low - order harmonics efficiently , no significant yield of high harmonics can be expected even for very high laser intensities .
this is a consequence of dynamical stochasticity , inherent to nonlinear dynamical systems driven by weak , time - dependent forces such as the screened electric field inside a cluster .
increasing the laser intensity does not help much because the self - consistent field trapping the electrons inside the ion core increases too . as a consequence the electron population always splits into deeply bound electrons and a halo , the mie - frequency and the screening increases , so that the physical picture remains almost insensitive to the intensity of the applied field .
another option we did not consider above is to use relatively long pulses where the first - order resonance with the mie frequency can be reached because of the ion core expansion .
if the mie resonance is met the ac electric field inside the cluster is grossly enhanced and nearly all electron trajectories appear to be strongly disturbed . in this case an analysis of near - resonant stochastic behavior as given above is inappropriate .
it seems that a direct numerical study is the only option under these conditions .
an analysis based on classical vlasov simulations was performed in refs.[14 ] .
we are grateful to w. becker for valuable discussions .
this work was supported by the deutsche forschungsgemeinschaft and the russian foundation for basic research ( project no.06 - 02 - 04006 ) .
t. dppner , th .
fennel , th .
diederich , j. tiggesbumker , and k.h .
meiwes - broer , , 013401 ( 2005 ) ; t. dppner , th .
fennel , p. radcliffe , j. tiggesbumker , and k.h .
meiwes - broer , , 031202(r ) ( 2006 ) ; th .
fennel , t. dppner , j. passig , ch .
schaal , j. tiggesbumker , and k.h .
meiwes - broer , , 143401 ( 2007 ) .
c. vozzi et al
. m. nisoli , j - p .
caumes , g. sansone , s. stagira , and s. de silvestri , m. vecchiocattivi , d. bassi , m. pascolini , l. poletto , p. villoresi , and g. tondello , appl .
* 86 * , 111121 ( 2005 ) . | harmonic emission from cluster nanoplasmas subject to short intense infrared laser pulses is studied . in a previous publication [ m. kundu _ et al .
_ , , 033201 ( 2007 ) ] we reported particle - in - cell simulation results showing resonant enhancements of low - order harmonics when the mie plasma frequency of the ionizing and expanding cluster resonates with the respective harmonic frequency .
simultaneously we found that high - order harmonics were barely present in the spectrum , even at high intensities .
the current paper is focused on the analytical modeling of the process .
we show that dynamical stochasticity owing to nonlinear resonance inhibits the emission of high order harmonics . |
according to the currently most favored cosmological model , the @xmath0cdm model ( @xmath0 denotes the cosmological constant , cdm stands for cold dark matter ) , our observable universe is geometrically flat and normal " , i.e. baryonic , matter is only responsible for about 4 - 5@xmath2 of its content .
most of the universe is composed of two exotic fluids " : dark matter ( dm ) and dark energy ( de ) .
neither of these exotic components can be observed directly so far , but there are arguments that support their existence .
dm explains galaxy rotation curves and plays a crucial role in cosmic structure formation .
de is seen as an effective fluid with negative pressure , possibly associated with the quantum vacuum .
this fluid " is able to account for the accelerated expansion of the scale factor of the robertson - walker metric of the standard model . even being a model with observational success ,
the @xmath0cdm model leaves open the physical nature of these dark components .
moreover , there are a number of tensions which still await clarification ( see , e.g. , @xcite ) .
there is therefore ongoing interest in discussing and probing alternative approaches which differ in one or more aspects from the @xmath0cdm dynamics .
one line of research does no longer regard dm and de to be independent components with separate energy - momentum conservation but investigates the consequences of a more complex structure of the dark sector , modeled by an interaction between its principal constituents .
admitting the possibility of a non - gravitational coupling between dm and de is , of course , the more general case which gives rise to a richer cosmological dynamics . in particular , this generalization implies the existence of de perturbations , even for cases of an equation - of - state ( eos ) parameter @xmath3 , sometimes called decaying vacuum models @xcite .
it has been pointed out that ignoring de perturbations may result in unreliable conclusions concerning the interpretation of observational data @xcite . for further studies about clustering de see , e.g. ,
@xcite . ignoring a potential interaction between both dark components altogether , may result in an incorrect interpretation of cosmological data @xcite .
many models have been established by now which consider different types of interactions @xcite .
some more recent studies include @xcite .
all of them are phenomenological since the physical nature of the dark sector largely remains a matter of speculation . in most cases interactions
are assumed to be linear in the sense that the coupling terms in the energy - momentum balances of the components are proportional either to the dm density or to the de density or to a linear combination of both densities ( see , e.g. , @xcite ) .
there exists a large body of literature which studies such models , usually resulting in limits on the type or the strength of the interaction .
now , from a physical point of view it seems more natural to prefer an interaction that depends on the product of the abundances of the individual components , as , e.g. , in chemical reactions .
it was shown in @xcite that a coupling proportional to the product of the densities of dm and de is also observationally favored over linear models .
further recent studies of non - linearly interacting de models are @xcite and @xcite .
systems with non - linear interactions do not allow , in most cases , an analytic treatment , not even for the homogeneous and isotropic background .
here we consider the special case of a non - linear interaction for which there exists an analytic background solution .
while the existence of such analytic solution is important in itself , we emphasize its usefulness in setting up the system of equations for the perturbation dynamics about this background .
this solution will be characterized by a single additional parameter the value of which serves to quantify and to restrict potential differences from the @xmath0cdm model in a simple and transparent way . using a suitable modification of the camb code @xcite we focus on the implications of the coupling on the anisotropy spectrum of the cosmic microwave background ( cmb ) and on the ( total ) matter power spectrum .
this complements and extends previous work on the status of the non - linear model using large - scale - structure data @xcite . in this context
we also point out that the simple use of the position of the first acoustic peak in the cmb anisotropy spectrum , something which was occasionally done in the literature , is not a reliable criterion in assessing competing cosmological models .
the structure of the paper is as follows . in sec .
[ gencos ] we present the basic set of fluid dynamical equations both for the background and for the first - order perturbation dynamics of a multi - component cosmic medium .
this system is applied to our model with non - linear interactions in the dark sector in sec .
[ scaling ] . the subsequent sec .
[ numeric ] is devoted to the numerical and statistical analysis of our model .
[ discussion ] summarizes and discusses our results .
we consider a spatially flat universe with a perturbed robertson - walker ( rw ) metric which reads , up to first order in the perturbations and in the synchronous gauge , @xmath4 . \label{metric}\ ] ] here , @xmath5 is the scale factor and @xmath6 is the conformal time . restricting ourselves to scalar perturbations , the first - order quantity @xmath7 has two scalars degrees of freedom @xmath8 and @xmath9 according to , @xmath10\right\rbrace d^{3}k .
\label{hij}\ ] ] we assume that the cosmic substratum is composed of a set of components that may or may not interact with each other . each component , characterized by a subindex @xmath11 , will be treated as a perfect fluid with an energy - momentum tensor @xmath12 where @xmath13 is the @xmath11-component s four - velocity and @xmath14 is a projection tensor orthogonal to the four - velocity . @xmath15 and @xmath16 are energy density and pressure , respectively , of component @xmath11 .
quantities without subindex @xmath11 will refer to the total fluid . like the metric tensor , the other dynamical quantities can be split into a sum of a zeroth - order term ( denoted by a bar ) and a first - order term ( denoted by a hat ) . since zeroth - order quantities live in a homogeneous and isotropic background , the dynamical quantities are @xmath17 in the last relation , the spatial perturbation of the four - velocity was written as a divergency of a scalar function , known as peculiar velocity potential .
the total energy - momentum tensor is the sum of all the individual energy - momentum tensors .
if a certain component interacts with one or more of the remaining components , its energy - momentum balance is affected by an interaction term @xmath18 , @xmath19 since we assume that gravity is described by general relativity ( gr ) , the total energy - momentum tensor must be conserved , i.e , the interaction terms must satisfy the relation @xmath20 the interaction term is split in a similar way as the quantities in equation ( [ quantities ] ) , @xmath21 following @xcite , is convenient to decompose this interaction @xmath22 into two terms : a term parallel to the four - velocity , and a term orthogonal to the four - velocity , @xmath23 the temporal component of the interaction term is associated to the energy transfer , and its spatial component to the transfer of momentum .
spatial homogeneity implies that the spatial component of the background interaction term must be identically zero . under these conditions the background dynamics
is described by friedmann s equations , @xmath24 while the first - order perturbation equations are @xmath25 where the prime denotes the derivative with respect to the conformal time , @xmath26 is the hubble parameter with respect to the conformal time and @xmath27 is a quantity that can be associated with the anisotropic stress .
the energy balance for each of the components can be obtained by the projection of the covariant derivative of the energy - momentum tensor on the four - velocity . in zeroth order
we assume the rest frames of the components to coincide .
for the zeroth - order energy balances we have @xmath28 where @xmath29 is the background equation - of - state ( eos ) parameter @xmath30 the first - order equation is @xmath31 the momentum conservation is obtained by the projection of the covariant derivative orthogonal to the four - velocity .
as already mentioned , the background momentum balance is identically zero , thus , the momentum contributes only in first order .
it reads @xmath32^{\prime}+4\mathcal{h}\left(\bar{\rho}_{x}+\bar{p}_{x}\right ) v_{x}+\hat{p}_{x}=a\left(\bar{q}_{x}v_{x}+\hat{f}_{x}\right ) .
\label{momentumgeneral}\ ] ] here , the scalar function @xmath33 was introduced via , @xmath34 the adiabatic sound speed is defined by @xmath35 to characterize the propagation of perturbations we introduce the comoving sound speed @xmath36 where @xmath37 and @xmath38 are the gauge - invariant comoving perturbations of pressure and energy density , respectively : @xmath39 an alternative way to write the pressure perturbations is @xmath40 the density contrast of component x will be described by @xmath41 now , using the definitions ( [ w ] ) , ( [ csadiabatic ] ) , ( [ cscomoving ] ) and ( [ deltageral ] ) , the first - order energy balance ( [ energygeneral ] ) in the @xmath42-space becomes @xmath43 , \label{kenergy}\end{aligned}\ ] ] while the momentum balance ( [ momentumgeneral ] ) yields , @xmath44-\dfrac{a } { \bar{\rho}_{x}}\hat{f}_{x}. & & \label{kmomentum}\end{aligned}\ ] ] these equations hold separately for each component .
we shall model now the content of the universe as a four - component fluid , consisting of radiation ( index @xmath45 ) , baryonic matter ( index @xmath46 ) , cdm ( index @xmath47 ) and de ( index @xmath0 ) .
we assume that there is an interaction between the dark components with ( cf .
( [ bianchi ] ) ) @xmath48 while baryons and radiation behave in the same way as in the @xmath0cdm model .
neither baryons nor photons couple directly to de and dm . they interact with each other via thomson scattering before recombination .
thus , the equations for these two components are the well - established boltzmann equations of .
for the dark sector we use the fluid eqs .
( [ kenergy ] ) and ( [ kmomentum ] ) since a microscopic description of de and dm with corresponding boltzmann equations is not yet available .
now we apply equations ( [ kenergy ] ) and ( [ kmomentum ] ) to the cdm and de components . for a constant eos parameter of the de component
this parameter coincides with the adiabatic sound speed . under this condition
the de energy balance becomes @xmath49 \label{deenergypert}\end{aligned}\ ] ] and the momentum balance reduces to @xmath50-\dfrac{a } { \bar{\rho}_{\lambda}}\hat{f}. & & \label{demomentumpert}\end{aligned}\ ] ] the analysis of the de dynamics has to be performed separately for the cases @xmath51 and @xmath52 . equation ( [ demomentumpert ] )
shows that for @xmath51 the de peculiar velocity is not a dynamic variable .
then , this equation can be used to determine @xmath53 explicitly in a model - independent way , @xmath54 this result will be used later in the cdm equation . for @xmath51
the de energy balance takes the form @xmath55 . \label{deenergypert1}\ ] ] in the second case,@xmath56 , the de peculiar velocity is dynamically relevant and equation ( [ demomentumpert ] ) must be solved to determine the temporal evolution of @xmath57 .
there is no way to determine the perturbation of the spatial interaction term as in the previous case .
therefore , we restrict ourselves to @xmath58 . the energy balance is again given by equation ( [ deenergypert ] ) , while the momentum balance simplifies to @xmath59 .
& & \label{demomentumpert1}\end{aligned}\ ] ] the cdm component is characterized by @xmath60 with negligible pressure perturbations . using the definitions ( [ csadiabatic ] ) , ( [ pcomoving ] ) and ( [ cscomoving ] )
one concludes that @xmath61 .
its perturbative energy balance is @xmath62 the perturbative cdm momentum balance is affected by the @xmath53 term . for @xmath51 use of ( [ hatf ] ) provides us with @xmath63 where @xmath53 is given by equation ( [ hatf ] ) . in the second case with @xmath64
the perturbative momentum balance reduces to @xmath65 our relations so far are valid for any cosmological model consisting of an effective de fluid which may interact with pressureless dm as well as of baryons and radiation , both the latter being treated as in the standard model . in the following
we shall consider a specific configuration which is analytically solvable in the homogeneous and isotropic background and can be understood as the consequence of a coupling between dm and de that is proportional to the product of the energy densities of both dark components .
our model of interest in this paper relies on a scaling behavior of the ratio of the energy densities of dm and de .
it can be demonstrated that it is this dynamics which is generated by an interaction term proportional to the product of the densities of dm and de .
following @xcite , we start by introducing a covariant length scale @xmath66 by @xcite @xmath67 we are looking for a class of models for which the ratio of the energy densities of cdm and de obeys a power law with respect to this length scale , @xmath68 these models are characterized by the free parameter @xmath69 . note that in a homogeneous and isotropic friedmann - lematre - robertson - walker ( flrw ) space - time the expansion scalar is equal to three times the hubble parameter and the length scale @xmath66 coincides with the scale factor @xmath5 of the rw metric .
relation ( [ scalinggeneral ] ) covariantly generalizes the model by dalal et al .
@xcite which was restricted to the background dynamics . different properties of this model and its observational consequences have been investigated in @xcite .
based on the covariant generalization ( [ scalinggeneral ] ) we are able to study the perturbation dynamics of the scaling model .
the @xmath0cdm model is recovered as the special case @xmath70 , @xmath51 , and @xmath71 .
the _ ansatz _ ( [ scalinggeneral ] ) induces an interaction between cdm and de .
according to ( [ bgenergy1 ] ) with ( [ q1dark ] ) , their background energy balances take the form @xmath72 combining the equations ( [ scalinggeneral ] ) , ( [ bgscalingcdm ] ) and ( [ bgscalingde ] ) and solving for the background interaction term yields @xmath73 a non - linear interaction of this type induces a scaling ( [ scalinggeneral ] ) of the ratio of the energy densities .
this coupling is proportional to the product of the energy densities of the interacting components which we consider to be more
realistic " than most of the other interacting models with interactions just linear in the energy density of the components .
it is only in the limit @xmath74 , i.e. at high redshift , that one approaches a linear dependence @xmath75 which is the preferred model in many studies in the field @xcite .
obviously , the interaction vanishes for @xmath76 . integrating equations ( [ bgscalingcdm ] ) and ( [ bgscalingde ] )
, we obtain the analytic solutions @xmath77 the friedmann equation ( [ friedmann1 ] ) then provides us with an expression for the hubble rate , @xmath78 . \label{scalinghubble}\ ] ] the present value @xmath79 of the hubble rate is , as usual , conveniently parametrized by @xmath80 . with ( [ scalinghubble ] ) the background dynamics
is explicitly known .
obviously , the existence of an analytic expression for the background hubble rate will be useful in dealing with the perturbation dynamics . to solve the first - order dynamics , an explicit expression for the perturbed interaction
is required . as already mentioned , the spatial term has the model independent form ( [ hatf ] ) .
for the temporal term @xmath81 we assume that the background relation ( [ bgscalingq0 ] ) is valid in general , thus on the perturbed level , @xmath82 , \label{hqscaling}\ ] ] where , following @xcite , the first - order expansion scalar is , @xmath83 the complete set of equations for the dark - sector perturbations is obtained by inserting ( [ hqscaling ] ) for @xmath81 into the first - order balance equations . for the case
@xmath51 we combine ( [ hqscaling ] ) and ( [ htheta ] ) with the equations ( [ deenergypert1 ] ) , ( [ cdmenergypert ] ) and ( [ cdmmomentumpert ] ) , for the case @xmath56 the expressions ( [ hqscaling ] ) and ( [ htheta ] ) are combined with equations ( [ deenergypert ] ) , ( [ demomentumpert1 ] ) , ( [ cdmenergypert ] ) and ( [ cdmmomentumpert1 ] ) .
many papers in the literature neglect the perturbations of de and of the interaction term , taking @xmath84 . here
we take into account these perturbations and we shall quantify their impact on the cmb power spectrum within a statistical analysis .
the formalism described above can be used now to test the scaling cosmology model against the observational data .
the main motivation of this chapter is to compare the parameter selection obtained by a background test using sne ia data from the jla sample @xcite , with a perturbative test using the cmb power spectrum data from planck @xcite . because of computational limitations , we restricted our analysis to the case @xmath51 .
in both cases we use a @xmath85 analysis , based on @xmath86 where @xmath87 , @xmath88 corresponds to the free parameters , @xmath89 denotes the covariance matrix of data @xmath90 and @xmath91 represents the theoretical predictions for a given set of parameters .
since we are modifying only the dark sector , we have @xmath92 .
moreover , in the cmb spectrum analysis we fix the set of parameters @xmath93 , where @xmath6 is the optical depth , @xmath94 is the initial perturbation amplitude and @xmath95 is the spectral index , using the best fit value for tt+low p+lensing results in @xcite . in the entire analysis ,
the hubble rate @xmath8 is left free but at the end it is marginalized in order to obtain @xmath96 contour curves .
as is well known , sne ia tests are using the luminosity distance modulus @xmath97+\mu_{0 } \label{musn}\ ] ] with @xmath98 , where @xmath99 is the luminosity distance @xmath100 here we extend a previous analysis @xcite by including a separately conserved baryon component using the hubble parameter ( [ scalinghubble ] ) .
this confirms that the sne ia analysis does not satisfactorily constrain the interaction parameter @xmath69 , i.e. , the luminosity distance modulus is not very sensitive to this parameter .
the results of our analysis are shown in fig . 1 . [ fig1 ] in order to obtain the cmb power spectrum for the scaling cosmology model , we modify the camb code by implementing the perturbation equations ( [ deenergypert1 ] ) , ( [ cdmenergypert ] ) , ( [ cdmmomentumpert ] ) , ( [ hqscaling ] ) and ( [ htheta ] ) .
to integrate the equations , the initial conditions were fixed following @xcite , which are valid in the non - adiabatic case and for more general sound speeds .
it turns out that the cmb spectrum is much more sensitive to the interaction parameter than the sne ia data .
using camb recursively , we found the values for @xmath101 and @xmath69 that best fit the _
planck _ data
. the values for @xmath69 are considerably stronger constrained than by the jla data .
small deviations from the @xmath0cdm model are still admitted which is in accordance with conclusions based on large - scale structure data @xcite .
the upper panel of fig . 2 shows the cmb spectrum for different values of the interaction parameter @xmath69 . for values of @xmath69 substantially different from the @xmath0cdm value
@xmath102 the cmb spectrum completely disagrees with the data even though the position of the first peak may be correct .
the lower panel of fig . 2 shows the contour curves of the @xmath69-@xmath101 plane based on the _ planck _ data and a summary of our studies is presented in table 1 .
the degeneracy of the parameter @xmath69 in the sne ia analysis is largely broken by the cmb analysis .
this becomes very evident in the superposition of contour curves in fig .
3 . fig2a + [ fig2b ] [ tab ] test & @xmath101 & @xmath8 & @xmath69 + sne ia ( jla ) & @xmath103 & @xmath104 & @xmath105 + cmb ( planck ) & @xmath106 & @xmath107 & @xmath108 + [ fig3 ] and 2@xmath27 ) for the combined cmb and snia analysis .
use of the cmb data drastically reduces the degeneracy in the parameter @xmath69.,title="fig:",scaledwidth=65.0% ] [ ] different from the @xmath0cdm model , any dynamical de model is necessarily equipped with inhomogeneities not only of the matter distribution but also of the de component itself .
our basic set of perturbation equations allows us to quantify the potential relevance of these de perturbations . to this purpose
we compare the cmb power spectrum for the best fit values of @xmath101 , @xmath8 and @xmath69 for the general case with @xmath109 , @xmath110 and @xmath111 ( our analysis so far ) with a simplified version of the model with an interaction only in the background , assuming vanishing de perturbations , i.e. , @xmath112 . as fig .
4 shows , the power spectra for both cases differ only very slightly at very large scales . on smaller scales which here also include the scale of the first acoustic peak
, we have identical results with and without de perturbations .
this means , apart from the integrated sachs - wolfe effect , the essential features of the cmb remain unaffected . for practical purposes
this may justify the use of a model with vanishing de perturbations .
the point is , however , that such a statement is possible only after the corresponding calculation has been done .
[ fig4 ] , @xmath69 and @xmath8 ) .
the best - fit values are @xmath113 , @xmath114 and @xmath115 .
the case where the interaction is only taken into account in the background equations and @xmath112 is almost identical with the general case with @xmath109 , @xmath110 and @xmath111 except at the largest scales .
, title="fig:",scaledwidth=90.0% ] [ ] our analysis of the cmb perturbations constrains @xmath69 to stay close to the @xmath0cdm value @xmath71 .
this complements the results of an earlier study based on large - scale structure data @xcite . here ,
the uncertainties in the mass variance at @xmath116 at the @xmath117 scale ( @xmath118 ) or in the initial amplitude of scalar perturbations prevent us from getting further constraints from a statistical analysis of the matter power spectrum which could improve those already obtained by the cmb analysis .
5 shows the dependence of the matter power spectrum on @xmath69 for @xmath51 .
differences from @xmath71 result in increasing or decreasing powers on large scales together with corresponding distortions of the shape at bao scales which do not match the large - scale structure observations .
[ fig5 ] for different values of @xmath69 .
the case @xmath71 corresponds to the @xmath0cdm model.,scaledwidth=65.0% ]
our main result is the calculation of the cmb spectrum in dependence of the model parameter @xmath69 where any value @xmath119 represents a deviation from the standard @xmath0cdm model due to a non - linear , non - gravitational interaction between dm and de . on the basis of the planck data we find the best - fit values @xmath114 and @xmath113 which coincide with the standard model at the 2@xmath27 confidence level .
the cmb data allow us to constrain the parameter @xmath69 with considerably higher precision than the snia data of the jla sample do . in this context
we clarified that the mere position of the first peak is not a suitable criterion for assessing a model .
perturbations of the de component where shown to be negligible , except , perhaps , on extremely large scales .
because of uncertainties in the mass variance or in the initial perturbation amplitude an analysis of the matter power spectrum does not improve the cmb based limits on @xmath69 .
+ * acknowledgement : * financial support by capes , fapes and cnpq ( brazil ) is gratefully acknowledged .
wshr was supported by fapes ( bpc no476/2013 ) at the begining of this work .
this work has made use of the computing facilities of national center for supercomputing ( cesup / ufrgs ) , and of the laboratory of astroinformatics ( iag / usp , nat / unicsul ) , whose purchase was made possible by the brazilian agency fapesp ( 2009/54006 - 4 ) and the inct - a .
ellis , gen .
. grav . * 41 * , 581 ( 2009 ) , ( reprint of g.f.r .
ellis in : r.k .
sachs ( ed . ) , proceedings of the international school of physics enrico fermi " ,
course 47 : general relativity and cosmology , pp .
104 - 182 . academic press , new york and london ( 1971 ) . ) | an interaction between dark matter and dark energy , proportional to the product of their energy densities , results in a scaling behavior of the ratio of these densities with respect to the scale factor of the robertson - walker metric .
this gives rise to a class of cosmological models which deviate from the standard model in an analytically tractable way .
in particular , it becomes possible to quantify the role of potential dark - energy perturbations .
we investigate the impact of this interaction on the structure formation process . using the ( modified ) camb code
we obtain the cmb spectrum as well as the linear matter power spectrum .
it is shown that the strong degeneracy in the parameter space present in the background analysis is considerably reduced by considering _ planck
_ data .
our analysis is compatible with the @xmath0cdm model at the @xmath1 confidence level with a slightly preferred direction of the energy flow from dark matter to dark energy . |
the existence of space - time singularities is a common property of most solutions of the general theory of relativity , including various black - hole or cosmological ones .
the well - known raychaudhuri equation @xcite has played a key role in describing such singularities @xcite .
this equation has also been used in different other contexts @xcite .
it has also attracted some attention beyond the classical descriptions , for example , a bohmian quantum version of the equation was recently introduced in ref.@xcite . in the present work
, we aim to investigate this equation in the framework of the weyl conformal theory of gravity by considering time - like world - lines .
this theory emerged from attempts to unify gravity with electromagnetism and since its advent in about a century ago , it continued to attract the attention of researchers in the field .
sharing several important solutions with the einstein theory of gravity , it also possesses the interesting property of being invariant under conformal transformations , see ref .
@xcite and references therein .
the relationships between this theory and the general theory of relativity have been discussed in several places , namely , in ref .
@xcite , in which maldacena has shown that it is possible to single out the einstein solution from the numerous solution to weyl theory by deploying a simple neumann boundary condition , and in ref .
@xcite , where it has been shown that varying the full connection of the weyl gravity results in the vacuum einstein equations .
null trajectories in weyl gravity have been studied in @xcite .
the connections between weyl gravity and extensions of the critical gravity was considered in @xcite . some difficulties with weyl gravity
have been discussed in refs .
@xcite . here
, we consider the vacuum solution of the weyl gravity which is a three - parameter schwarzschild - like solution supplemented by linear and quadratic terms .
this solution has been used in different proposals , say , in @xcite to explain the galactic rotation curves , in @xcite to study the sagnac effect , in @xcite to investigate strong lensing , and in @xcite to study gravito - magnetic effects .
the classical issues of bending of light and perihelion precession have been re - examined with this theory in refs . @xcite and @xcite , respectively . our motivations for the present study originates from both the interest in raychaudhuri equation and its applications in different contexts , and the weyl conformal gravity as a theory of gravity with conformal invariance .
the conformal invariance in gravity theories is regarded as an essential symmetry , see e.g. the discussion in @xcite .
also , modified gravities with quadratic curvature have attracted a lot of attention in recent decade and weyl gravity as a particular model in this class , deserves further study in this regard .
in what follows , we start with the action and field equations of the weyl conformal gravity .
we then proceed by a brief review of the vacuum static solution of the theory and relevant equations for time - like geodesics . using these equations ,
we apply the raychaudhuri equation to a congruence of radial flow and a flow with both radial and rotational velocities to investigate the focusing of geodesics . after comparing the results with those of the standard schwarzschild space - time , we investigate the effect of conformal transformations .
a summary of the results will conclude the work .
the weyl theory of gravity is a theory of fourth order with respect to the metric .
it is characterized by the action @xmath0 where @xmath1 is the weyl invariant , and @xmath2 is a coupling constant . using the properties of the gauss - bonnet invariant ,
the above action can be rewritten as @xmath3 varying the action given in eq .
( [ 11 ] ) , supplemented by a matter action , with respect to @xmath4 one gets the following field equation @xmath5 in which @xmath6 is the bach tensor @xcite , and @xmath7 is the energy - momentum tensor .
the vacuum field equation @xmath8 admits a static spherically symmetric solution defined by the line element @xmath9 in which @xmath10 this solution was first introduced in ref .
@xcite . by choosing appropriate values for the parameters @xmath11 , and @xmath12 , the schwarzschild - de sitter metric could be regenerated .
the parameters @xmath13 and @xmath12 can be related to the mass of the source and the cosmological constant , respectively .
in fact , @xmath14 gives the source mass .
the third parameter , @xmath15 is thus the crucial one here , carrying additional physical content , which might be related to dark matter @xcite .
for the the space - time described by the metric ( [ 14 ] ) and ( [ 15 ] ) , if we take @xmath16 , then depending on the sign of @xmath15 , there are one or two horizons corresponding to the roots of @xmath17 .
these are given by @xmath18 by assuming @xmath19 , we obtain @xmath20 which are consistent if @xmath21 .
if , in addition , @xmath22 , then @xmath23 . for @xmath24 , the second root is ruled out , and one is left with the horizon corresponding to @xmath25 . the function @xmath17 is plotted in fig .
[ fig0 ] for both typical positive and negative values of @xmath15 .
as this figure shows , in the case where @xmath21 , there is a maximum whose location is given by @xmath26 , and the maximum value is given by @xmath27 taking @xmath28 , these have the numerical values @xmath29 ; and @xmath30 for @xmath31 , respectively . in terms of @xmath32 for @xmath28 ; @xmath33 ( lower ) , @xmath34 ( middle ) , and @xmath35 ( upper ) . ]
now , we study the evolution of time - like geodesics in the space - time defined by eq .
( [ 14 ] ) .
we start with the 4-velocity vector field @xmath36 with @xmath37 , which defines a time - like congruence in the space - time under consideration . here , @xmath38 stands for @xmath39 , @xmath40 being an affine parameter along the relevant integral curves . due to the symmetries of the space - time described by eq .
( [ 15 ] ) , we can set @xmath41 .
then , the time - like killing vectors @xmath42 and @xmath43 together with the fact that @xmath44 ( see , e.g. , @xcite ) , result in @xmath45 where @xmath46 are constants .
thus , from the geodesic equation @xmath47 or , from the normalization of the four - velocity , one obtains @xmath48 for a radial flow , @xmath49 , we have @xmath50 in schwarzschild space - time , for @xmath51 , we obtain @xmath52 .
thus , by setting @xmath53 , one can choose @xmath54 , as viewed by an observer at infinity .
in fact , it is well - known that @xmath55 is equal to the particle energy in units of @xmath56 . in the space - time under consideration here , which is not asymptotically flat , we can not simply start by choosing @xmath53 . to fix the constant @xmath55 , one needs to use some initial conditions , say , the initial value of @xmath57 , or the maximum distance the particles can reach .
the choice of the value of @xmath55 affects the motion of particles .
this can be seen from eq .
( [ 24 ] ) .
for , @xmath49 and @xmath21 , if we choose @xmath58 , there will be turning points whose locations , @xmath59 , are given by the solutions of the equation @xmath60 , or equivalently , @xmath61 , and the particles motion are confined to @xmath62 , or @xmath63 , depending on the initial positions .
if we choose @xmath64 , there will be no turning points and particles can pass through @xmath65
. it would be useful to note that @xmath66 is equal to , for example , about @xmath67 for @xmath31 , respectively .
we now use the well - known raychaudhuri equation @xmath68 to study the behavior of the above congruence . here , @xmath69 , @xmath70 , and @xmath71}$ ] are the scalar expansion , the symmetric traceless shear tensor , the antisymmetric rotation tensor , respectively , @xmath72 , @xmath73 , and @xmath74 . for the radial congruence described by eq .
( [ 24 ] ) , it can be shown that the rotation tensor vanishes and therefore , @xmath75 the right hand side of this equation is negative for @xmath76 , which means focusing of the congruence . however , for @xmath21 , the right hand side is not necessarily negative and the equation does not guarantee convergence . to analyze the latter case
, we use a direct calculation of the expansion . by neglecting the cosmological constant
, we obtain @xmath77 in which the negative sign corresponds to outgoing world - lines @xmath78 , and the positive sign is associated with incoming ones with @xmath79 .
now , the rate of change of this expression can be obtained either by direct differentiation , or by using the right hand side of eq .
( [ 27 ] ) .
the resulting expression have a rather algebraically complicated numerator quartic in @xmath32 , and factors of @xmath80 in the denominator .
thus , for an arbitrary value of @xmath55 , there are up to four sign changes ( corresponding to the roots of the numerator ) , and a divergency ( in either side of @xmath65 , corresponding to the roots of the expression in the denominator ) . for clarity , we consider the case where there are no turning points , which is the case for @xmath64 . for convenience , we take @xmath81 which means @xmath82 . the function @xmath83 is plotted in fig .
[ fig4 ] in terms of @xmath32 and @xmath84 for an incoming congruence with a typical value @xmath85 .
this shows that the world - line are convergent . in terms of @xmath32 and @xmath84 for @xmath86 .
] to have a better understanding of the physical content of eq .
( [ 26 ] ) , we consider the galaxy cluster abell 370 which is known particularly in relation with strong lensing phenomenon @xcite .
it has a mass of @xmath87 . using this
, we may set @xmath88 .
thus , we have @xmath89 , @xmath90 , i.e. , the linear term and @xmath91 term in @xmath17 are of the same order . also , @xmath92 , while other terms ( except @xmath13 ) in the numerator of @xmath93 in eq .
( [ 26 ] ) are some orders of magnitude smaller , that is , the main contribution to @xmath93 comes from the first two terms in the numerator .
pure rotational flow is also possible with zero expansion as for schwarzschild space - time . for combined radial - rotational flow ( in the equatorial plane )
, we have from eq .
( [ 23 ] ) @xmath94 in which @xmath95 one can show that the relevant rotation tensor vanishes for this flow , and the raychaudhuri equation reduces to @xmath96 again , leading to convergent congruence for non - negative @xmath15 . for negative @xmath15 , the above equation does not result in convergence necessarily .
in fact , in this case we have @xmath97 where @xmath98
this results in a rather complicated expression for @xmath83 , and general behavior of the congruence depends on both @xmath46 . for radial null geodesics , the geodesic equation results in @xmath99 by choosing @xmath53 , the raychaudhuri equation reads @xmath100 that is , focusing with no signature of @xmath15 .
this has the solution @xmath101 .
the action for weyl gravity and the associated vacuum field equation are invariant under the conformal transformation @xmath102 where @xmath103 is a scalar field .
thus , any metric related to the vacuum solution of the field equations via conformal transformations is also a solution .
however , time - like geodesics are not necessarily mapped to geodesics .
thus , one can ask to what extent results obtained from equations like raychaudhuri equation in a given space - time are valid in the transformed space - time . in this section
, we aim to investigate the behavior of time - like congruences and the raychaudhuri equation under such transformation .
we start with @xmath104 and @xmath105 from which it can be shown that @xmath106 for null geodesics , the last term in the right - hand side of the above equation disappears and one is left with the geodesic equation written in non - affine form .
this is a re - statement of the well - known fact that null geodesics remain geodesics under a conformal transformation . for time - like geodesics ,
this is not the case .
regarding the raychaudhuri equation for time - like congruences , we have @xmath107 or @xmath108 this means that , in general , there is no guarantee that a convergent congruence remains convergent in the transformed space - time . for a scalar field
@xmath109 which depends only on the radial coordinate , eq .
( [ 13f ] ) reduces to @xmath110 where a prime stands for differentiation with respect to the radial coordinate . as an example , if we choose @xmath111 , we obtain and @xmath112 , which leads to function @xmath113 plotted in fig .
it shows that the congruence is convergent in this particular transformed space . in terms of @xmath32 for , @xmath111 , @xmath28 , @xmath114
the above results show how the trajectories behave in the transformed space .
it is also possible to have a form of raychaudhuri equation unchanged when transforming to a conformally related space .
this can be reached by considering a dynamical particle mass @xcite .
in conclusion , we studied the focusing of time - like congruence within the space - time described by the static vacuum solution to the weyl conformal gravity model . for galactic scales , in which we ignored the cosmological constant ,
a term proportional to @xmath115 is added to the right - hand side of the raychaudhuri equation , where the rotation term is already absent . for positive @xmath15
this ensures the convergence of incoming geodesics with a rate greater than the case of the schwarzschild space - time .
we also obtained a relation between the rate of convergence of a congruence for a given solution and in its conformally transformed space - time .
the resultant equations do not guarantee that a convergent congruence remains so in the transformed space . in the context of cosmology
, these results might help to study how cosmic flows will behave in weyl conformal gravity .
we would like to thank payame noor university for financial support . 0 a.k .
raychaudhuri , _ phys .
rev . _ * 98 * , 1123 ( 1955 ) .
r. penrose , _ phys .
* 14 * , 57 ( 1965 ) .
s.w . hawking , r. penrose , _ proc .
a _ * 314 * , 529 ( 1970 ) .
albareti , j.a.r .
cembranos , a. de la cruz - dombriz , _ jcap _ * 12 * , 020 ( 2012 ) .
s. rippl , h. van elst , r. tavakol , d. taylor , _ gen .
_ * 28 * , 193 ( 1996 ) .
s. kar , _ phys .
d _ * 54 * , 6408 ( 1996 ) .
f. shojai , a. shojai , _ phys.rev .
d _ * 78 * , 104011 ( 2008 ) .
t. harko , f.s.n .
lobo , _ phys .
* 86 * , 124034 ( 2012 ) .
a.p . kouretsis .
m. stathakopoulos , p.c .
stavrinos , _ phys .
d _ * 86 * , 124025 ( 2012 ) . c. valls , _ comm .
phys . _ * 308 * , 133 ( 2011 ) .
j. borgman , l.h .
ford , _ phys .
d _ * 70 * , 064032 ( 2004 ) .
stavrinos , _ gen .
grav . _ * 44 * , 3029 ( 2012 ) .
tsagas , m.i .
kadiltzolglou , _ phys .
d _ * 88 * , 083501 ( 2013 ) .
albareti , j.a.r .
cembranos , a. de la cruz - dombriz , a. dobado , _ jcap _ * 07 * , 009 ( 2013 ) .
tsagas , _ class .
quantum grav . _
* 23 * , 4323 ( 2006 ) .
kouretsis , c.g .
tsagas , _ phys .
d _ * 82 * , 124053 ( 2010 ) .
m. mohseni , _ gen .
grav . _ * 47 * , 24 ( 2015 ) .
s. das , _ phys .
* 89 * , 084068 ( 2014 ) .
mannheim , d. kazanas , _ astrophys . j. _ * 342 * , 635 ( 1989 ) .
j. maldacena , _ arxiv_:1105.5632 ( 2011 ) .
wheeler , _ phys .
d _ * 90 * , 025027 ( 2014 ) .
villanueva , m. olivares , _ jcap _ * 1306 * , 040 ( 2013 ) .
h. lu , y. pang and c.n .
pope , _ phys .
d _ * 84 * , 064001 ( 2011 ) .
flanagan , _ phys .
d _ * 74 * , 023002 ( 2006 ) .
y. yoon , _ phys .
d _ * 88 * , 027504 ( 2013 ) .
j. sultana , _ gen .
* 46 * , 1710 ( 2014 ) .
d. cutajar , k. zarb adami , _ mon . not .
r. astron .
soc . _ * 441 * , 1291 ( 2014 ) .
said , j. sultana , k. zarb adami , _ phys .
d _ * 88 * , 087504 ( 2013 ) .
j. sultana , d. kazanas , _ phys .
d _ * 81 * , 127502 ( 2010 ) .
s. pireaux , _ class .
quantum grav . _
* 21 * , 1897 ( 2004 ) .
j. sultana , d. kazanas , j.l . said , _ phys .
d _ * 86 * , 084008 ( 2012 ) .
g. t hooft , _
_ [ gr - qc ] ( 2014 ) .
carroll , _ arxiv _ : gr - qc/9712019 ( 1997 ) .
j. richard , j .-
kneib , m. limousin , a. edge , e. jullo , _ mon . not .
r. astron .
soc . _ * 402 * , l44 ( 2010 ) .
mannheim , _ gen .
grav . _ * 25 * , 697 ( 1993 ) . | we study the evolution of time - like congruences in the vacuum solutions of weyl conformal theory of gravity . using the raycaudhuri equation , we show that for positive values of the coefficient of the linear term in the solution and in the absence of the cosmological constant , the incoming rays converge .
the evolution of the congruence for negative values is investigated for different values of the parameters .
the behavior of the congruence under conformal transformations is also studied .
+ pacs : 04.20cv , 04.50kd |
the low - temperature collective physics of interacting quantum many - body systems often calls for a novel description in terms of emergent degrees of freedom that are not only distinct from those of the original constituents of the system , but describe certain `` fractions '' thereof .
familiar examples include the spin - charge separation in one - dimensional metals @xcite , the electron fractionalization in fractional quantum hall states of two - dimensional electron gases @xcite , as well as the emergence of monopoles in spin ice @xcite or chiral magnets @xcite .
quantum spin liquids in frustrated quantum magnets @xcite provide another important venue for such quantum number fractionalization .
for these spin liquids the theoretical formulation of this phenomenon is often closely linked to a lattice gauge theory description of the quantum magnet ; the spin degrees of freedom typically decompose into spinons coupled to an emergent @xmath1 or z@xmath0 gauge field whose elementary excitations remain deconfined @xcite .
one of the paradigmatic examples of a model harboring a z@xmath0 spin liquid ground state is kitaev s exactly solvable honeycomb model @xcite .
it describes a spin-@xmath2 quantum magnet subject to strong exchange frustration arising from bond - directional interactions of the form @xmath3 where @xmath4 labels the three different bond directions of the honeycomb lattice .
the low - energy physics of this spin model can be captured in terms of majorana degrees of freedom and a z@xmath0 gauge field .
crucially , the gauge field remains static for the pure kitaev model , and identifying the ground state configuration of the gauge field reduces to an essentially classical problem .
typically this yields a unique ground state with a finite gap for the elementary vison excitations of the z@xmath0 gauge field .
fixing the gauge structure then allows to recast the original spin model as a free majorana fermion model and thus paves the way to a full analytical solution .
the phase diagram of the kitaev model generically exhibits two types of spin - liquid phases . around the limits where one of the three couplings dominates over the other two one finds a gapped spin liquid which , for the two - dimensional honeycomb model , is known to exhibit abelian topological order @xcite . the second phase , which is found for roughly isotropic couplings ( i.e. , @xmath5 ) is gapless and can generically be understood as a metallic state of the itinerant majorana fermions . for the two - dimensional honeycomb model the itinerant majorana fermions form a graphene - like band structure with two dirac cones @xcite .
l13 mm | c70 mm c15 mm c20 mm c20 mm c c lattice & alternative & sites in & sublattice & inversion & + & names & unit cell & symmetry & symmetry & symbol & no .
+ ( 10,3)a & hyperoctagon @xcite , laves graph @xcite , k@xmath6 crystal @xcite & 4 & @xmath7 & chiral & i@xmath8 & 214 + ( 10,3)b & hyperhoneycomb @xcite & 4 & & & fddd & 70 + ( 10,3)c & & 6 & & chiral & p@xmath9 & 151 + ( 9,3)a & & 12 & & & r@xmath10 m & 166 + ( 9,3)b & & 24 & & & p4@xmath11/nmc & 137 + ( 8,3)a & & 6 & @xmath7 & chiral & p@xmath12 & 180 + ( 8,3)b & & 6 & @xmath7 & & r@xmath10 m & 166 + ( 8,3)c & & 8 & & & p@xmath13 / mmc & 194 + ( 8,3)n & & 16 & & @xmath14 & i4 / mmm & 139 + ( 6,3 ) & honeycomb & 2 & & & & in this paper , we comprehensively classify the nature of the gapless spin liquids and their underlying majorana metals for _ three - dimensional _ kitaev models .
our motivation has been rooted in the mounting experimental evidence that spin - orbit entangled mott insulators can provide solid - state realizations of the kitaev model following the theoretical guidance by khaliullin and coworkers @xcite .
this materials - oriented search @xcite has produced various candidate 4d and 5d compounds , most notably na@xmath15iro@xmath16 , @xmath17-li@xmath15iro@xmath16 and rucl@xmath16 , which realize hexagonal arrangements of local , spin - orbit entangled @xmath18 moments that are indeed subject to strong bond - directional exchanges as indicated by recent experiments @xcite .
a byproduct of this experimental search has been the discovery @xcite of the polymorphs @xmath19-li@xmath15iro@xmath16 and @xmath20-li@xmath15iro@xmath16 , which realize three - dimensional arrangements of the spin - orbit entangled moments which retain the _ tricoordination _ familiar from the hexagonal lattice .
this has sparked a surge of interest in three - dimensional variants of the kitaev model which , hitherto , had evaded the attention of the broader community @xcite .
it was quickly recognized that the analytical tractability of the two - dimensional kitaev model largely carries over to the three - dimensional variants , and it has recently been demonstrated that such three - dimensional kitaev models harbor a rich variety of gapless z@xmath0 spin liquids in which the emergent majorana metals form nodal structures which include majorana fermi surfaces @xcite , nodal lines @xcite as well as topologically protected weyl nodes @xcite . the purpose of this paper is to go beyond these initial examples and to impart a more systematic classification of gapless kitaev spin liquids in three spatial dimensions . in particular , we comprehensively discuss how the nature of the emergent majorana metal depends on the underlying lattice geometry .
we do so by considering kitaev models for the most elementary three - dimensional , tricoordinated lattices , i.e. , lattices that have elementary loops of only one fixed length @xcite . for instance , the well - known honeycomb lattice is the only tricoordinated lattice with elementary loops of length 6 .
however , there are multiple lattice structures with elementary loops of lengths 7 , 8 , 9 or 10 ( and possibly higher ) , which are all three - dimensional .
in fact , such three - dimensional , tricoordinated structures have been comprehensively classified in the work of wells in the 1970 s @xcite . here , we focus on those lattice structures that exhibit equidistant bonds and approximately 120@xmath21 bond angles at every vertex @xcite .
an overview of the so - identified family of three dimensional , tricoordinated lattice structures and their basic properties is provided in table [ tab : lattice_overview ] .
a convenient way to systematically label the individual lattices is to use the so - called schlfli symbol @xmath22 followed by a letter , where @xmath23 is the fixed polygonality ( or elementary loop length ) of the lattice , @xmath24 refers to the tricoordination of the vertices , and the additional letter simply enumerates the lattices for a given schlfli symbol .
it should be noted that some of these lattices are well known in the literature under alternative names .
these include the ( 10,3)a lattice , which has long been known as the laves graph @xcite in the crystallographic literature or as @xmath25 crystal @xcite in the mathematical literature .
it has also been renamed hyperoctagon lattice @xcite by some of the authors of this manuscript in an earlier study .
similarly , the ( 10,3)b lattice has recently gained some attention under the name hyperhoneycomb lattice @xcite after it had been discovered as the iridium - sublattice in the iridate @xmath19-li@xmath15iro@xmath16 .
it is precisely this family of tricoordinated lattice structures that serves as principal input in our quest to comprehensively discuss three - dimensional kitaev models in the following .
we show that these kitaev models harbor a plethora of gapless spin liquids that can be cast as different incarnations of majorana metals whose precise nature can be systematically understood from a basic symmetry analysis .
l13 mm | c3 cm | c3 cm lattice & majorana metal & trs breaking + ( 10,3)a & fermi surface & fermi surface + ( 10,3)b & nodal line & weyl nodes + ( 10,3)c & nodal line & fermi surface + ( 9,3)a@xmath26 & weyl nodes & weyl nodes + ( 8,3)a & fermi surface & fermi surface + ( 8,3)b & weyl nodes & weyl nodes + ( 8,3)c@xmath26 & nodal line & weyl nodes + ( 8,3)n & gapped & weyl nodes + ( 6,3 ) & dirac cones & gapped before going into a detailed discussion of the kitaev models for these individual lattices , we provide a brief overview of our main results . for all but one lattice , i.e. ,
( 8,3)n , we find that there is an extended gapless spin - liquid phase around the point of isotropic coupling , i.e. , @xmath5 .
this gapless phase is best described as a majorana metal ( or semimetal ) , since it is the band structure of the itinerant majorana fermions that exhibits gapless excitations , while the vison excitations of the static z@xmath0 gauge field remain gapped for all lattices @xcite .
a summary of our results characterizing the various majorana metals for different lattice geometries is provided in table [ tab : majorana_metals ] .
we do so by listing the nodal manifold of gapless excitations in the band structure of the itinerant majorana fermions .
as can be seen from the table , the various three - dimensional lattice geometries realize multiple examples for the emergence of fermi surfaces , nodal lines , or weyl nodes . as we will lay out in detail in the remainder of the paper , an understanding of the systematics in this table
is closely linked to a symmetry analysis of the kitaev models for the respective lattice geometries ( see in particular sec . [
sec : symmetries ] ) .
for instance , the occurrence of majorana fermi surfaces for the two lattices ( 10,3)a and ( 8,3)a is closely linked to a non trivial sublattice symmetry for these lattices .
similarly , the emergence of weyl nodes can be understood from a close inspection of time - reversal and inversion symmetry .
for instance , for a lattice with an _ odd _ number of bonds in the elementary loop , such as the ( 9,3)a lattice , time - reversal symmetry has to be broken spontaneously for the emergent majorana degrees of freedom , which in turn allows for the occurrence of weyl nodes . for the lattice ( 8,3)b it is a more intricate interplay of time - reversal symmetry , inversion symmetry , and non - trivial sublattice symmetry that allows for the emergence of weyl nodes in the majorana band structure without breaking time - reversal symmetry _ nor _ inversion symmetry , a situation that can not occur for electronic band structures .
we will discuss these different incarnations of weyl physics in a broader context in sec .
[ sec : weyl ] .
for the two - dimensional kitaev honeycomb model , it is well known that the dirac spin - liquid of the pure kitaev model gaps out into a massive phase with non - abelian topological order when perturbing the spin system with a magnetic field pointing along the 111-direction , i.e. , by considering a hamiltonian @xmath27 it may , thus , be a natural question to ask whether the breaking of time - reversal symmetry can give rise to gapped phases with non - trivial topological order also for three - dimensional kitaev models .
bearing in mind that the kitaev model can be recast as a free majorana fermion system , one can immediately answer this question in the negative by considering the classification of topological insulators @xcite rooted in the symmetry classification of free - fermion systems @xcite .
in this free - fermion classification scheme , the pure kitaev model falls into symmetry class bdi , while the one with broken time - reversal symmetry of eq
. belongs to symmetry class d. as can readily be seen from the classification tables of refs .
, there are no topologically non - trivial band insulators in either symmetry class bdi or d for three spatial dimensions in contrast to two spatial dimensions where symmetry class d allows for this possibility [ and as realized for hamiltonian on the honeycomb lattice ] .
it is , of course , nevertheless an interesting question to ask what effect the breaking of time - reversal symmetry has on the majorana metals for the three - dimensional kitaev models of table [ tab : majorana_metals ] .
the answer is provided in the third column of that table and is discussed in much detail in the remainder of the paper .
generically , we find that the metallic nature per se remains untouched by the breaking of time - reversal symmetry , i.e. , for no system do we observe a transition into a ( topologically trivial ) massive phase . instead , we find that the nodal structure remains robust under this perturbation for systems that exhibit majorana fermi surfaces or weyl nodes
. the only effect of a ( small ) magnetic field is to either deform the majorana fermi surface or move the positions of the weyl nodes .
a different picture emerges for those lattices where the pure kitaev model exhibits nodal lines . here
, the magnetic field does alter the nodal structure .
for the lattices ( 10,3)b and ( 8,3)c the magnetic field gaps out the nodal line with the exception of two or six nodes , respectively .
these nodes turn out to be one or three pairs of weyl nodes . a symmetry analysis again indicates that another symmetry plays a crucial role in stabilizing this unusual band structure .
it is the presence of inversion symmetry that fixes these weyl nodes to the fermi energy ( at zero energy ) for these lattices .
inversion symmetry is absent for the chiral lattice ( 10,3)c , which also exhibits nodal lines for the unperturbed kitaev model . upon applying a magnetic field
the nodal lines vanish and the system creates six pairs of weyl nodes , which in the absence of inversion symmetry are no longer fixed to the fermi energy and move away from it . the result is the emergence of 12 pockets of majorana fermi surfaces , each of which encapsulates a weyl node . as a result , these fermi surfaces also acquire some non - trivial topological properties from the weyl nodes , as we discuss in further detail in sec .
[ ssec : topfermisurface ] . as such , the kitaev model for lattice ( 10,3)c stands out as the only system where the effect of breaking time - reversal symmetry is to _ increase _ the nodal structure and the associated density of states of the majorana metal .
finally , we briefly comment on the z@xmath0 gauge structure of these models . as we will discuss in more detail in sec . [ sec : kitaev ] , the assignment of local z@xmath0 gauges ( on the bonds of the lattice ) corresponds to an assignment of z@xmath0 fluxes through the elementary loops in the lattice .
for the two - dimensional kitaev honeycomb model , this correspondence can be used to readily fix the gauge structure of the ground state via a theorem by lieb @xcite , which states that the ground state of the honeycomb model has no fluxes through any of the hexagonal plaquettes ( corresponding to the elementary loops ) .
lieb s theorem can also be applied to three - dimensional kitaev models if the lattice structure exhibits certain mirror symmetries .
as it turns out , only one of the lattices in our family , namely the lattice ( 8,3)b , fulfills this criterion .
it is thus the only lattice for which we can rigorously assign the flux configuration of the ground state . for all other lattices
, we have to resort to alternative ways to identify precisely this ground state configuration of the ( static ) z@xmath0 gauge field . in this study , we have resorted to numerical simulations of this essentially classical problem for finite systems . in general , we find that the result of this numerical procedure is a flux configuration that precisely corresponds to the one indicated by lieb s theorem if it were to apply to the lattice structure at hand .
specifically , all elementary loops of length 2 ( mod 4 ) carry zero flux , while elementary loops of length 0 ( mod 4 ) carry @xmath28 flux .
non - bipartite lattices with elementary loops of odd length are not covered by lieb s theorem at all . for these lattices
we assign the flux configuration using symmetry arguments .
notably , however , for the lattice ( 9,3)a our numerical checks indicate the possibility of low - energy flux assignments that break at least one of the point - group symmetries of the lattice .
such an exotic scenario might also be relevant to lattice ( 8,3)c where the @xmath28-fluxes are subject to geometric frustration .
we have not explored this possibility in full detail in the paper at hand , but instead have provided results for the lowest - energy z@xmath0 flux structure that does not break any point - group symmetries .
we revisit this point in an outlook at the end of the paper .
the paper is structured as follows . in sec .
[ sec : kitaev ] , we briefly review the general framework to analytically solve the kitaev model in arbitrary spatial dimensions . a detailed analysis of the relevant symmetries of three - dimensional kitaev models is presented in sec . [
sec : symmetries ] .
we then go through all lattices of table [ tab : lattice_overview ] one by one in sec .
[ sec:3d ] and provide further details on the underlying lattices , the definition of the kitaev model , and its solution for each of these lattices .
this includes the overall phase diagram of the model along with a detailed discussion of the gapless phase around the point of isotropic couplings . in the subsequent sec .
[ sec : weyl ] , we take a step back and discuss the different scenarios for the emergence of weyl physics in these three - dimensional kitaev models .
this also includes a discussion of the topological properties of some of the observed majorana fermi surfaces .
section [ sec : spin - peierls ] focuses on a discussion of majorana fermi surfaces in general and their bcs - type spin - peierls instabilities .
we round off the paper with an outlook in sec .
[ sec : outlook ] that touches on the possibility of realizing some of the kitaev models of interest here in spin - orbit entangled mott insulators .
we further lay out some future directions to be pursued for this family of three - dimensional kitaev models .
the main paper is complemented by an extensive appendix that provides many of the technical details on the lattice structures .
we start our discussion by briefly reviewing the main traits of kitaev s original solution @xcite of the honeycomb model and lay out how it can be adapted to the three - dimensional model systems of interest here . at the heart of kitaev s exact solution
is the existence of a macroscopic number of conserved quantities , which are associated with closed loops in the underlying two- or three - dimensional lattice . for each of these loops
, we can define a corresponding loop operator by @xmath29 where @xmath30 labels the sites within the loop @xmath31 and @xmath32 is given by @xmath33 for even - length loops , the operator has eigenvalue @xmath34 , where eigenvalue @xmath35 is identified with the presence of a z@xmath0 flux through the enclosed plaquette and eigenvalue @xmath36 implies that there is no flux through the plaquette .
if the operator contains an odd number of bonds / sites , its eigenvalue is instead @xmath37 .
this case is treated in more detail in the discussion of lattice ( 9,3)a in sec .
[ ssec:9a ] .
note that the definition is chosen in order to be consistent with the conventions chosen by lieb in discussing his flux - theorem @xcite for the ground state . .
] the set of loop operators is generically linearly dependent for three - dimensional ( 3d ) lattices , due to the presence of volume constraints as exemplified in fig .
[ fig : minimalvolumes ] .
in particular , if a set of loop operators forms the boundary of a closed volume , the product of their eigenvalues is fixed to @xmath36 .
thus , we are going to restrict the discussion to a linearly independent subset of these .
for each of the lattices considered here , we can identify @xmath38 fundamental loop operators ( per unit cell ) , where @xmath39 is the number of sites per unit cell , from which all other loop operators can be built by successive multiplication .
the hilbert space factorizes into eigenstates of these loop operators ( in the following called `` flux sectors '' ) and we will usually restrict our analysis to a single such flux sector .
another consequence of the volume constraint is that flux excitations , also called visons , always form closed loops ; there are no magnetic monopoles in the corresponding z@xmath0 gauge field .
following kitaev s original solution @xcite , we proceed to represent each spin in terms of four majorana fermion operators @xmath40 with @xmath41 , and @xmath42 denoting the site within the unit cell at position @xmath43 .
the majorana fermion operators obey the usual anti - commutation relations @xmath44 this representation faithfully reproduces the spin algebra within the physical hilbert space defined by @xmath45 as a second step , we regroup the majoranas into bond operators @xmath46 , with @xmath20 being the label of the nearest - neighbor bond @xmath47 .
the bond operators have eigenvalues @xmath34 , which therefore can be identified with an emergent z@xmath0 gauge field .
note that , in contrast to the loop operators , these z@xmath0 gauge fields are _ not _ physical , but merely a consequence of enlarging the hilbert space in eq . .
in particular , we can identify the loop operators as the gauge - invariant quantities of this emergent z@xmath0 gauge field .
it turns out that gauge transformations play an important role when classifying the possible majorana metals , as they affect how symmetries are implemented in the majorana system .
this is discussed in more detail in sec .
[ sec : symmetries ] .
l13 mm | c15 mm c15 mm c15 mm c15 mm lattice & flux sector & lieb & vison gap & vison loop + & & theorem & & length + ( 10,3)a & @xmath48-flux & & 0.09(1 ) & 10 + ( 10,3)b & @xmath48-flux & & 0.13(1 ) & 6 + ( 10,3)c & @xmath48-flux & & 0.13(1 ) & 3 + ( 9,3)a@xmath26 & @xmath49-fluxes & & & 4 + ( 8,3)a & @xmath28-flux & & 0.07(1 ) & 2 + ( 8,3)b & @xmath28-flux & yes & 0.16(1 ) & 2 + ( 8,3)c@xmath26 & @xmath48-flux & & & 4 + ( 8,3)n & @xmath28-flux & & 0.16(1 ) & 2 + ( 6,3 ) & @xmath48-flux & yes & 0.27 & 2 it can easily be shown that all bond operators commute with each other , as well as with all of the loop operators and with the hamiltonian .
thus , we can fix their eigenvalues @xmath50 and solve the resulting quadratic hamiltonian for the @xmath51 majoranas @xmath52 note that when assigning eigenvalues to the bond operators , we need to pick a directionality as @xmath53 @xcite .
the remaining difficulty is to decide on how to assign the bond eigenvalues .
one guiding principle is lieb s theorem @xcite .
this theorem determines the ground - state flux configuration for any plaquette that is invariant under mirror symmetry , as long as the mirror line ( 2d ) or plane ( 3d ) does not cut through lattice sites , but only bonds .
the flux per plaquettes is 0 if the loop length equals 2 ( mod 4 ) , whereas it is @xmath28 for loop lengths 0 ( mod 4 ) . for the honeycomb lattice
, we show the relevant mirror line in fig .
[ fig : lieb ] ( a ) .
the bond length 6 implies that the ground state sector has 0 flux per plaquette , i.e. all loop operators have eigenvalue @xmath36 .
rotations around the @xmath54-direction .
the combination determines the eigenvalues of all the fundamental loop operators .
( c ) one of the mirror planes for ( 8,3)n .
another is obtained by a 90@xmath55 rotation around @xmath54 . a third mirror plane ( not shown )
lies in the @xmath56-plane .
these determine the eigenvalues of all but one loop operator per unit cell . ] for most of the 3d lattices discussed here , lieb s theorem can not be applied .
notable exceptions are the lattices ( 8,3)b and ( 8,3)n . for the former ,
lieb s theorem determines the flux of all the fundamental loop operators ; for the latter , lieb s theorem determines seven of the eight fundamental loop operators per unit cell .
examples of the relevant mirror planes are shown in fig .
[ fig : lieb ] ( b ) for the lattice ( 8,3)b and in fig .
[ fig : lieb ] ( c ) for ( 8,3)n .
for all the other lattices , one needs to resort to numerical simulations to determine the flux configuration of the ground state .
however , lieb s theorem still provides a good educated guess . in particular , choosing the flux configuration as explained above and preserving the symmetries of the underlying lattices often yields the correct ground state sector .
notable exceptions are ( 9,3)a and ( 8,3)c , where numerical studies indicate that the system may prefer to spontaneously break lattice symmetries in the ground state . . ]
note that the elementary excitations of the z@xmath0 gauge field remain gapped for all lattices .
we calculated this vison gap for all lattices [ but lattices ( 8,3)c and ( 9,3)a , see the discussion above ] for the smallest possible vison loop .
this can usually be obtained by flipping a single @xmath57-type bond , except for the lattices ( 10,3)b and c , where instead a single @xmath58- or @xmath59-type bond is flipped .
results for the vison gap are given in table [ tab : z2 ] and fig .
[ fig : visongap ] , where the explicit system - size dependence of the vison gap is illustrated .
details on the smallest vison loop and which bond is flipped can be found in appendix [ sec : supplemental ] .
[ [ magnetic - field . ] ] magnetic field. + + + + + + + + + + + + + + + + when discussing the effect of symmetries on possible majorana metals , we are particularly interested in the effect of time reversal and what consequences breaking time - reversal symmetry can have . for simplicity , the time - reversal symmetry breaking term we consider is chosen as an external magnetic field in the 111 direction .
the particular direction is not essential as long as it couples to all spin components , but it ensures that all other ( lattice ) symmetries remain intact . as @xmath60 does not commute with the loop operators ,
the model is no longer exactly solvable . however , as long as the visons are gapped and the strength of @xmath61 sufficiently small as not to excite them , we can treat the magnetic field term perturbatively . following ref .
@xcite , one finds that the first non - trivial contribution arises in third - order perturbation theory and can be written as a three - spin interaction , @xmath62 where the summation is over all triples of adjacent sites @xmath42 , @xmath63 , and @xmath64 such that the bond @xmath65 ( @xmath66 ) is of @xmath17- ( @xmath19- ) type and @xmath20 is chosen as the remaining bond type .
the prefactor @xmath67 depends on the vison gap @xmath68 and the strength of the magnetic field .
using the representation in terms of majorana fermions , we can rewrite this as an effective next - nearest neighbor hopping of the @xmath51 majoranas by @xmath69 where we can neglect the operator @xmath70 as it will act as the identity on the physical subspace .
while the general framework of kitaev s original solution of the honeycomb model can be applied to arbitrary tricoordinated lattices , including three - dimensional ones , important differences between individual lattice structures arise when considering the most elementary symmetries of the system . in particular , it is important to note that the symmetries of the original spin hamiltonian can manifest themselves in distinct ways when considering their effect in the physical majorana subspace .
these projective symmetries turn out to be sensitive to the underlying lattice geometry .
the deeper origin of this lattice dependence can be traced back to the somewhat subtle incarnation of the z@xmath0 gauge theory description of the kitaev model .
all calculations are done in a fixed gauge , i.e. , one chooses a specific set of bond operator eigenvalues that is compatible with the flux sector . in order for the symmetries to act within this fixed gauge sector
, they often need to be supplemented by a gauge transformation .
as we detail in the following , such a supplemental gauge transformation may lead to an additional shift in momentum space in the projective symmetry relations for the majorana fermions . as we will argue in the following ,
a careful analysis of these projective symmetries , in particular particle - hole , time - reversal , and inversion symmetry , will allow us to systematically classify the kitaev hamiltonian in its majorana representation and infer the nature of the emergent majorana metal from it .
[ [ particle - hole - symmetry - phs . ] ] particle - hole symmetry _
( phs)_. + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + particle - hole symmetry is , strictly speaking , not a symmetry of our systems , but rather a consequence of describing the spins in terms of majorana fermions ( instead of complex fermions ) which over - counts the degrees of freedom by a factor of 2 .
the majorana condition @xmath71 immediately leads to the condition @xmath72 in momentum space . for the hamiltonian and its eigenenergies , this implies @xmath73 where the asterisk indicates complex conjugation .
the over - counting of degrees of freedom can be taken care of in two ways : either by restricting the allowed momenta to half the brillouin zone or by discarding the lower half of the energy bands . .
] [ [ sublattice - symmetry - sls . ] ] sublattice symmetry _ ( sls)_. + + + + + + + + + + + + + + + + + + + + + + + + + + + + +
all lattices that are considered here , with the exception of ( 9,3)a , are bipartite lattices , i.e. , the lattice sites can be partitioned into two sublattices ( referred to as sublattices a and b in the following ) such that nearest neighbors are always from different sublattices .
the pure kitaev hamiltonian has only nearest neighbor interactions .
consequently , a sublattice transformation , @xmath74 changes the overall sign of the hamiltonian and implies that @xmath75 where @xmath76 is a unitary matrix and @xmath77 is a reciprocal lattice vector of the sublattice .
note that @xmath78 , if the sublattice has the same translation vectors as the full lattice .
examples for this case are the honeycomb lattice in 2d and the ( 10,3)b ( hyperhoneycomb ) lattice in 3d . if the translation vectors connect the two sublattices , as is the case , e.g. , for the square lattice in fig .
[ fig : sublattice ] , then @xmath77 has a finite value ; in particular , @xmath79 for the square lattice . in 3d , @xmath77 is non - vanishing for the lattices ( 10,3)a ( hyperoctagon ) , ( 8,3)a , as well as for ( 8,3)b . [
[ time - reversal - symmetry - trs . ] ] time - reversal symmetry _
( trs)_. + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + the importance of the sublattice symmetry becomes apparent when trying to implement time - reversal symmetry for the majorana system .
as in the kitaev honeycomb model @xcite , we consider a time - reversal transformation @xmath80 that squares to one , @xmath81 , which implies symmetry class bdi .
time - reversal symmetry flips the spin eigenvalues , and can be implemented in the majorana language by @xmath82 and complex conjugation . however , due to the complex conjugation it also flips the eigenvalue of all the bond operators , @xmath83 and , thus , needs to be supplemented by a gauge transformation so as to remain in the same ( fixed ) gauge sector .
the required gauge transformation is , in fact , simply the sublattice transformation discussed above , and we can implement time - reversal by requiring @xmath84 where @xmath85 or @xmath35 depending on whether the site is in the a- or b - sublattice .
thus , the gauge - invariant time - reversal @xmath86 inherits the @xmath77 vector from the sublattice transformation , and we obtain @xmath87 note that if @xmath78 , the only stable zero - mode manifolds are nodal lines independent of other symmetries , as was shown in @xcite . if @xmath7 or time - reversal symmetry is broken , then other symmetries become important for determining the stable zero - energy modes , see also table [ tab : majorana_metals ] .
[ [ inversion - symmetry - i . ] ] inversion symmetry _
( i)_. + + + + + + + + + + + + + + + + + + + + + + + + + + analogously to time - reversal invariance discussed above , inversion symmetry also needs to be supplemented by a gauge transformation in order to act within a fixed gauge sector .
however , the exact form of the gauge transformation depends on the details of the lattice and the flux configuration . in general ,
inversion symmetry will act as @xmath88 where @xmath89 is either half a reciprocal lattice vector or zero , depending on whether the necessary gauge transformation enlarges the unit cell or not .
in particular , @xmath90 may or may not be equal to @xmath77 , and various different possibilities are realized in the models we discuss here . of particular interest is the lattice ( 8,3)b , where @xmath7 and @xmath91 , thus allowing for the presence of weyl nodes with both inversion and time - reversal symmetries unbroken . an example for @xmath92 and @xmath93 is the lattice ( 8,3)n , where the time - reversal broken model allows for majorana fermi surfaces , even in the presence of inversion symmetry .
note that other lattice symmetries , such as rotations , may also have such an additional translation in momentum space .
we now turn to a detailed discussion of the 3d kitaev models for the various three - dimensional , tricoordinated lattices of table [ tab : lattice_overview ] . for each lattice , we go through a similar line of arguments where we ( i ) provide some elementary information about the lattice structure such as its unit cell , associated lattice vectors and the assignment of kitaev couplings to the bonds of the lattice , ( ii ) discuss the structure of the elementary loops and the assignment of z@xmath0 fluxes in the ground state of the kitaev model , ( iii ) determine the manifestation of the projective symmetries , and ( iv ) discuss the nature of the emergent majorana metal .
this discussion is supplemented by two appendices where we provide more detailed technical information for each lattice . in appendix
[ sec : appendix3dkitaev ] we present visualizations of the lattice structures along various high - symmetry projections along with detailed information on their space group and the wyckoff positions for the unit cell .
this lattice information is further supplemented by vesta @xcite visualization files , which are provided in the supplemental material of this paper . in appendix
[ sec : supplemental ] , we provide additional information on the 3d kitaev models for these lattices .
in particular , we provide a detailed summary of the gauge structure of the ground state and give an explicit expression for the kitaev hamiltonian in its majorana representation in this gauge .
are co - rotating spirals , while the ( 8,3)b lattice [ shown in ( b ) ] has counter - rotating spirals . the two different rotation directions are indicated by orange and blue , respectively . ]
we start the discussion with lattices of elementary loop length 8 . of the 15 nets reported on by wells @xcite , only four have equal - length bonds and 120@xmath21 bond angles and
will be discussed in detail in the coming sections .
the first two lattices , ( 8,3)a and ( 8,3)b , are in fact close cousins of each other . both of them can be viewed as a three - dimensional version of the 3 - 12 - 12 lattice @xcite or , alternatively , the yao - kivelson lattice @xcite , where the triangles are replaced by triangular spirals .
while these spirals are co - rotating in ( 8,3)a , resulting in a chiral lattice , they are counter - rotating in ( 8,3)b , thus leading to an inversion - symmetric lattice .
a comparison of the two lattices is given in fig .
[ fig : comparison_8_3_a_b ] , where the two different rotation directions are marked by different colors . and @xmath94 , as well as between @xmath95 and @xmath96 . ] [ [ lattice - structure . ] ] lattice structure.
+ + + + + + + + + + + + + + + + + + + more formally , the ( 8,3)a lattice is a hexagonal lattice with six sites per unit cell at positions @xmath97 we choose the lattice vectors as @xmath98 and the corresponding reciprocal lattice vectors become @xmath99 the unit cell and the translation vectors are illustrated in fig .
[ fig : phase_diagram_8_3a ] ( a ) .
the color coding of bonds in this figure indicates the assignment of the bond - directional coupling along @xmath58- , @xmath59- , and @xmath57-type bonds with colors green , red , and blue , respectively .
this particular assignment of the bonds is chosen such as to retain as many of the lattice symmetries as possible , and is _ unique _ up to an overall permutation of the three bond types .
note that the sets of all @xmath58- , @xmath59- , and @xmath57-bonds are related to one another by the threefold screw - rotation symmetry around the @xmath54 axis ; consequently , the phase diagram , shown in fig .
[ fig : phase_diagram_8_3a ] ( b ) , has to be symmetric in all couplings .
[ [ gauge - structure . ] ] gauge structure. + + + + + + + + + + + + + + + + + the ( 8,3)a lattice has three loop operators of length 8 and three of length 14 per unit cell .
these six loop operators form three closed volumes which leads to only three linearly independent loop operators per unit cell ( see appendix [ ssec : supplemental_8_a ] for details ) . in what follows
, we consider the flux sector where all loop operators of length 8 have eigenvalue @xmath35 and all loop operators of length 14 have eigenvalue @xmath36 .
this configuration of fluxes respects all the lattice symmetries and , although this lattice does not possess the symmetries required for rigorous application of lieb s theorem @xcite , is consistent with the flux assignments one would expect were lieb s theorem to hold .
in addition , we have also checked numerically that this flux sector is indeed the ground - state sector .
the vison gap for this lattice , reported in fig .
[ fig : visongap ] , is computed by flipping a single @xmath57-bond operator , which changes the signs of four loop operators . for further details ,
we refer the reader to appendix [ ssec : supplemental_8_a ] .
[ [ projective - symmetries . ] ] projective symmetries. + + + + + + + + + + + + + + + + + + + + + + + this lattice has the property that the translation vector @xmath100 maps the two sublattices onto each other .
therefore , sublattice symmetry and time - reversal symmetry involve a non - vanishing translation in momentum space by @xmath101 .
as the lattice is chiral , the relevant energy relations are given by particle - hole and time - reversal symmetry @xmath102 [ [ majorana - metal . ] ] majorana metal. + + + + + + + + + + + + + + + + following the symmetry analysis of sec .
[ sec : symmetries ] , the projective symmetries of lattice ( 8,3)a indicate that the only stable zero - energy manifolds are _
, we find an extended gapless phase around the point of isotropic coupling [ see the phase diagram shown in fig . [
fig : phase_diagram_8_3a ] ( b ) ] , where the gapless modes sit on four majorana fermi surfaces , visualized in fig .
[ fig : phase_diagram_8_3a ] ( c ) .
the darker shaded orange region of the phase diagram denotes the parameter space where these majorana fermi surfaces are topological , i.e. , they contain a weyl point at finite energy .
the weyl points can be seen in the energy dispersion in fig .
[ fig : energydispersion8a ] as the band crossings between the middle two bands . for isotropic coupling strengths ,
the weyl points are located at @xmath103 and their translations by @xmath77 , while pairs of ( oppositely charged ) weyl nodes are located at the touching points of the surfaces [ see fig . [
fig : phase_diagram_8_3a](c ) ] .
the latter are located at @xmath104 , and @xmath105 , as well as their translations by @xmath77 .
note that combining time - reversal with particle - hole symmetry implies that the spectrum is anti - symmetric under translation of @xmath77 , i.e. , @xmath106 [ assuming energies are sorted @xmath107 , which is clearly visible in the dispersion plot in fig .
[ fig : energydispersion8a ] . for a more detailed description on topological fermi surfaces , as well as the evolution of the position of weyl points throughout the phase diagram
, we refer the reader to the discussion on topological fermi surfaces in sec .
[ ssec : topfermisurface ] . as can be seen from fig .
[ fig : phase_diagram_8_3a ] ( c ) , pairs of majorana fermi surfaces are mapped onto each other by the perfect nesting vector @xmath77 .
this suggests a fermi surface instability as soon as the majorana fermions become interacting , which happens naturally when adding additional interactions to the pure kitaev hamiltonian .
a very similar situation occurs in the ( 10,3)a hyperoctagon lattice ( for details , see sec .
[ ssec:10a ] ) and was studied in ref .
it was shown that generic interactions always induce a bcs - type instability of the majorana fermi surface , albeit with the important difference that it is translation symmetry that is spontaneously broken , not @xmath1 symmetry .
thus , the instability was dubbed spin - peierls bcs instability .
the resulting phase is still a quantum spin liquid , but with a nodal line instead of a full fermi surface . as the arguments in ref .
@xcite are very general and only rely on the perfect nesting condition , we expect the same type of behavior for ( 8,3)a in the presence of interactions
. one important difference to the ( 10,3)a lattice lies in the larger number of surfaces for the ( 8,3)a lattice . as a consequence , time - reversal symmetry does not guarantee the presence of nodal lines , and interactions might thus gap the system completely . for further details on the spin - peierls instability , as well as the effect of time - reversal breaking interactions ,
the reader is referred to section [ sec : spin - peierls ] .
[ [ lattice - structure.-1 ] ] lattice structure. + + + + + + + + + + + + + + + + + + + the ( 8,3)b lattice has a lattice structure very similar to the lattice ( 8,3)a discussed in the previous section .
it can again be viewed as coupled triangular spirals , however , in contrast to ( 8,3)a , the rotation directions alternate between nearest - neighbor spirals in ( 8,3)b , leading to an inversion symmetric lattice .
it has six sites per unit cell that are located at @xmath108 and the lattice vectors are chosen as @xmath109 the corresponding reciprocal lattice vectors are given by @xmath110 the unit cell and translation vectors are illustrated in fig .
[ fig : phase_diagram_8_3b ] ( a ) .
the assignment of the three bond types is chosen analogously to ( 8,3)a ; again , this is the most symmetric assignment of the different bond types and unique up to overall permutations of @xmath58 , @xmath59 , and @xmath57 .
note that also for this lattice , the sets of all @xmath58- , @xmath59- , and @xmath57-bonds are related to one another by the threefold rotation symmetry around the @xmath54 axis , and the phase diagram , consequently , is symmetric in all couplings . [ [ gauge - structure.-1 ] ] gauge structure. + + + + + + + + + + + + + + + + + the lattice ( 8,3)b takes a special role among the lattices considered in this paper , as it is the only three - dimensional lattice that allows for a direct application of lieb s theorem @xcite to determine the ground - state flux sector _ rigorously_. the mirror planes used for determining the flux are illustrated in fig .
[ fig : lieb ] .
there are three linearly independent loop operators per unit cell , which all have length eight and , thus , @xmath28 flux in the ground state . flipping a single
@xmath57-type bond changes the sign of four plaquettes .
the corresponding vison gap @xmath111 ( measured in units of the kitaev coupling at the isotropic point @xmath112 ) is the largest found for the lattices considered here ( see table [ tab : z2 ] and fig .
[ fig : visongap ] ) .
rotation invariant line , marked in blue .
the right - hand side shows the energy dispersion along this high - symmetry line . ]
[ [ projective - symmetries - and - majorana - metal . ] ] projective symmetries and majorana metal. + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + this lattice has a non vanishing @xmath113 vector for time - reversal symmetry which , in the absence of other symmetries , would imply that the system exhibits stable majorana fermi surfaces .
however , as ( 8,3)b is also inversion symmetric with @xmath114 , the energy dispersion is particle - hole symmetric at every momentum which , in turn , prohibits stable fermi surfaces . instead , the system exhibits gapless _
weyl points _ in a finite parameter region around the isotropic point , as shown in the phase diagram of fig .
[ fig : phase_diagram_8_3b ] ( b ) and the dispersion plot of fig .
[ fig : dispersion_8_3b ] .
note that the weyl points are fixed to zero energy due to inversion symmetry . ) , are indicated as a guide to the eye . ] before discussing the manifestation of weyl physics in the 3d kitaev model for the ( 8,3)b lattice , we want to briefly recapitulate some elementary aspects of weyl semi - metals as they are typically discussed in the context of electronic band structures @xcite .
weyl points are , in fact , a very common phenomenon in three - dimensional band models , as band touching points generically show a linear dispersion . projected onto the two relevant bands , the low - energy band hamiltonian can ( to leading order )
be expanded as @xmath115 where @xmath116 denotes the displacement ( in momentum space ) from the band - touching point .
the `` velocities '' @xmath117 are , in general , all non - zero and linearly independent , in which case we call the band touching a weyl point ( wp ) .
each wp carries a charge , or chirality , that is defined by sgn@xmath118 $ ] .
more mathematically , we can identify wps with monopoles of `` berry flux '' , defined by @xmath119 where the summation in the berry connection @xmath120 runs over all occupied bands and @xmath121 denotes the bloch state of band @xmath122 at momentum @xmath123 .
as such , wps are topological objects and can only be gapped out in pairs of opposite chirality . in absence of scattering between the wps ,
this is only possible by bringing a pair of them with opposite chirality to the same point in momentum space , where they then can annihilate each other . throughout this paper
, we will mark weyl points of positive / negative chirality by red / yellow dots , respectively . due to the overall ( berry )
charge neutrality , weyl points always occur in pairs . for (
8,3)b , however , we find that weyl points need to occur in multiples of four as long as time - reversal remains intact . as particle - hole symmetry maps a weyl node at @xmath124 to a weyl node of _ opposite _ chirality at @xmath125 , while time - reversal maps it to a weyl node of _ the same _ chirality at @xmath126 .
thus , there are in total four weyl points , located at @xmath127 and @xmath128 . at the isotropic point , we find positively charged weyl points at @xmath129 and @xmath130 and negatively charged weyl points at @xmath131 and @xmath132 .
these four weyl points are visualized in fig .
[ fig : phase_diagram_8_3b](c ) . the charge or chirality of a wp can be measured by computing the chern number on an arbitrary closed surface ( in momentum space ) around it .
an alternative setup is computing the chern number on parallel planes and observing its jump when moving the plane `` through '' the wp . that these two methods give the same information on the wp ,
can readily be seen by noting that a closed surface can be smoothly deformed into a pair of planes , albeit with normal vectors that point in opposite directions .
reversing the normal vector of one of the planes is equivalent to changing its chern number , thus identifying the chern number of the closed surface to the _ difference _ in chern number of two planes on either side of the wp .
the latter method is slightly easier to implement numerically and will be the method of choice in this paper . for the lattice ( 8,3)b , the chern number as a function of @xmath133 is shown in fig .
[ fig : chern_numbers_8_3b ] .
the left - hand side shows the brillouin zone with the position of the weyl points and planes at positions @xmath134 , the right - hand - side shows the corresponding chern number . as a guide to the eye
, we marked the positions of the weyl points and the three example planes in the plot .
note that at each position of the weyl points , the chern number jumps by its respective charge .
figure [ fig : fermi_arcs_8_3b ] ( a ) shows the evolution of the weyl nodes in the 3d brillouin zone as the exchange couplings are varied with @xmath135 and @xmath136 . the position of weyl nodes of _ negative _ chirality is marked by colors changing from yellow to green as @xmath137 is increased , while weyl nodes of _ positive _ chirality are marked by colors changing from red to green . as @xmath137 is increased ,
weyl nodes of opposite chirality are seen to move towards each other , ultimately meeting and mutually annihilating for @xmath138 at @xmath139 and @xmath77 . for decreasing @xmath137
the weyl nodes move in the brillouin zone , but rather than meeting and annihilating at isolated momenta , the velocity vectors of the isolated weyl points approach zero , collapsing the bulk gap .
figure [ fig : fermi_arcs_8_3b ] ( b ) shows the associated fermi arc surface states in the 001-surface brillouin zone , visualized in fig .
[ fig : fermi_arcs_8_3b ] ( c ) , for a slab geometry .
fermi arcs are exact zero - energy surface modes that connect weyl points of opposite chirality .
similar to the weyl points themselves , the fermi arcs are also topologically protected . as long as the weyl points remain intact ,
no disorder or any other type of perturbation can gap these surface states .
as the weyl nodes move around , the fermi arcs are seen to deform , ultimately shrinking to nothing as the weyl nodes meet and annihilate for @xmath138 . for @xmath140 ,
the fermi arcs are seen to cross each other in the surface brillouin zone . as @xmath137 is increased further , the fermi arcs split once again .
while they still connect the same pairs of weyl nodes as before , they now wind differently around the brillouin zone .
this splitting / reconnecting of fermi arcs is purely a surface effect ( see appendix [ ssec : supplemental_8_b ] for details ) . breaking time - reversal symmetry with eq .
also causes the weyl nodes to wander around the brillouin zone .
figure [ fig : weyl_points_8_3b ] ( a ) shows the evolution of the weyl nodes for @xmath141 . at @xmath142 ,
two weyl nodes of opposite chirality meet and mutually annihilate at a high - symmetry point in the brillouin zone . in fig .
[ fig : weyl_points_8_3b ] ( b ) is pictured the corresponding evolution of the fermi arcs in the 001-surface brillouin zone .
as @xmath143 is increased from @xmath48 to @xmath144 , the fermi arcs become more warped as two weyl nodes of opposite chirality move towards each other . as @xmath143
is increased further , the two fermi arcs touch at a high - symmetry point and become one . for still larger values of @xmath143 , even more weyl nodes begin to appear in charge - neutral pairs while others pairs mutually annihilate . [ [ lattice - structure.-2 ] ] lattice structure. + + + + + + + + + + + + + + + + + + + the lattice ( 8,3)c can be viewed most simply as parallel zig - zag chains along the @xmath54 direction that are coupled by vertices lying in the @xmath58-@xmath59 plane . it is a hexagonal lattice with eight sites per unit cell at positions @xmath145 and lattice vectors @xmath146 the corresponding reciprocal lattice vectors are given by @xmath147 the unit cell and the lattice translation vectors are visualized in fig . [
fig : phase_diagram_8_3c](a ) .
when choosing the assignment of bond types on this lattice , we notice that for each of the chains there are two possible choices that are in general inequivalent .
we chose an assignment of bonds such that threefold rotation around the @xmath54 axis combined with a cyclic permutation of @xmath58,@xmath59 , and @xmath57 bonds is a symmetry of the hamiltonian .
this is the most natural choice as it ensures that the phase diagram remains symmetric under interchange of the @xmath148 , @xmath149 , and @xmath137 couplings . and @xmath150 ( gray ) .
the zero - energy surface immediately gaps completely when departing from the isotropic point . ]
[ [ gauge - structure.-2 ] ] gauge structure. + + + + + + + + + + + + + + + + + for this lattice there are four linearly independent loop operators of length 8 . following the guidance of lieb s theorem @xcite , one would like to assign @xmath28 flux through all of these plaquettes .
however , this is not possible as the minimal volume constraints on the loop operators indicate that only two out of three eight - length loops in such a volume can have @xmath28 flux ( as explained in more detail in appendix [ ssec : supplemental_8_c ] ) . as such
, the minimal volume constraints induce geometric frustration in assigning these @xmath28 fluxes , which gives rise to a macroscopic number of possible z@xmath0 gauge configurations .
while numerical tests indeed support that the ground state of the z@xmath0 gauge theory becomes macroscopically degenerate for increasing system sizes , we dismiss this scenario in the following discussion of the majorana physics . instead , we consider the flux sector where all loop operators have eigenvalue @xmath36 instead .
this turns out to be the only flux configuration that obeys all the lattice symmetries , i.e. , threefold rotation and inversion , but is not geometrically frustrated .
the reader should note that this flux configuration is _ not _ the ground state sector .
note , however , that one can always stabilize this flux sector as the ground state by adding terms that penalize @xmath28-flux through plaquettes , similarly to what was done in ref .
@xcite for the kitaev model on the two - dimensional square - octagon lattice .
[ [ projective - symmetries.-1 ] ] projective symmetries. + + + + + + + + + + + + + + + + + + + + + + + as the translation vectors of this lattice are identical to those of its two sublattices , sublattice symmetry and time - reversal symmetry are implemented trivially , i.e. , with @xmath78 .
in addition , this lattice possesses inversion symmetry with vanishing @xmath151 .
thus , the relevant energy relations are given by @xmath152 where the first relation comes from particle - hole symmetry and the second from either time - reversal or inversion symmetry .
[ [ majorana - metal.-1 ] ] majorana metal. + + + + + + + + + + + + + + + + the system exhibits a gapless phase around the isotropic point , where the gapless modes form closed _
nodal lines _ in accordance with our projective symmetry analysis . at the isotropic point , these lines are located at @xmath153 , as pictured in fig .
[ fig : phase_diagram_8_3c](c ) .
in addition , at the isotropic point one also finds a gapless surface at the top and bottom boundaries of the brillouin zone , @xmath154 .
in contrast to the nodal line , the surface is not a stable zero - mode manifold , as can readily be inferred from time - reversal invariance .
any infinitesimal changes in the coupling constants immediately gap the surface , while merely deforming the nodal line .
this becomes clearly visible in the energy dispersion plot along the high - symmetry lines , shown in fig .
[ fig : energydispersion8c ] .
the colored bands denote the energy dispersion at the isotropic point .
the zero - energy modes at the top and bottom surfaces of the brillouin zone gap out immediately when departing from the isotropic point , which is exemplified by the energy dispersion at @xmath155 and @xmath156 marked in gray . as discussed in sec .
[ sec : symmetries ] , the nodal line is protected by time - reversal symmetry .
thus , breaking time - reversal symmetry even infinitesimally causes the fermi line to gap out almost entirely , leaving just six weyl nodes which are fixed to zero energy by inversion symmetry .
a similar behavior was studied in ref .
@xcite for the lattice ( 10,3)b , discussed in sec .
[ ssec:10b ] .
the appearance of weyl points is far from being a coincidence .
in fact , it can be argued that time - reversal symmetry _ can not _ gap the nodal line completely and weyl nodes have to occur . a simple way to see this is by regarding the 3d model as a 2d model with an additional ( dimensional ) parameter , i.e. , we will view two of the momentum directions as physical and the third , for example , @xmath157 in this case , as a parameter that we can tune .
choosing @xmath157 such that the plane of physical momenta does not cut through the nodal line yields an effective two - dimensional system that is a trivial insulator @xcite .
if the plane of physical momenta instead _ does _ cut the line , the effective system will be gapless and exhibit two - dimensional dirac points , in complete analogy to the kitaev honeycomb model in the gapless phase .
breaking time - reversal symmetry has different effects for these two cases .
while it leaves the trivial insulator qualitatively unchanged , it induces a non - trivial gap for the dirac points and turns the effective system into a _ topological chern insulator_. in particular , the latter has non - vanishing berry flux through the two - dimensional brillouin zone .
thus , we found that changing the momentum parameter in absence of time - reversal symmetry , tunes the system between a trivial insulator and a chern insulator . from general arguments , we know that at the boundary between these two types of insulators , the system has to be gapless .
what is more , the gapless mode has to be a source / sink of berry flux in order to create the non - vanishing berry flux in the topological part
i.e. a weyl point . note that due to the periodicity of the brillouin zone , these weyl points always occur in pairs of opposite chirality . from @xmath158 .
( b ) corresponding fermi arc plot for @xmath159 .
( c ) visualization of the surface brillouin zone for open boundary conditions in the 100-direction . ]
as noted above , breaking time - reversal symmetry infinitesimally gaps the fermi line with the exception of six zero energy weyl nodes .
pictured in fig .
[ fig : weyl_points_8_3c ] ( a ) is the evolution of these weyl nodes for @xmath160 .
for this range of @xmath143 , the weyl points move very little in the brillouin zone along high - symmetry lines .
however , for larger values of @xmath143 , many weyl points appear in charge - neutral pairs while other pairs mutually annihilate .
figure [ fig : weyl_points_8_3c ] ( b ) shows the corresponding fermi arcs in the 100-surface brillouin zone for @xmath159 .
figure [ fig : weyl_points_8_3c ] ( c ) illustrates the projection of the weyl nodes onto the 100-surface brillouin zone .
finally , we turn to the ( 8,3)n lattice .
this lattice can be viewed as a three - dimensional generalization of the square - octagon lattice , where layers of square - octagon lattices are coupled via mid - bond sites .
[ [ lattice - structure.-3 ] ] lattice structure. + + + + + + + + + + + + + + + + + + + the ( 8,3)n lattice is a tetragonal lattice with 16 sites per unit cell . in order to simplify the notation
, we denote all vectors in terms of @xmath161 the site positions in the unit cell can be written as @xmath162 with @xmath163 and @xmath164 .
the lattice vectors are given by @xmath165 with reciprocal lattice vectors @xmath166 the unit cell and the lattice translation vectors are shown in fig .
[ fig : phase_diagram_8_3n](a ) .
the assignment of bonds was chosen in order to be compatible with fourfold rotation symmetry and inversion .
it is the unique such choice up to an overall permutation of the @xmath58 , @xmath59 , and @xmath57 bonds .
as can be seen from the figure , @xmath58- and @xmath59-bonds are related by lattice symmetries , but the @xmath57-bonds map only to themselves . as a result ,
the phase diagram is symmetric in interchanging @xmath148 and @xmath149 , only .
[ [ gauge - structure.-3 ] ] gauge structure. + + + + + + + + + + + + + + + + + the ( 8,3)n lattice has eight linearly independent loop operators per unit cell of lengths 8 and 10 , respectively . lieb s
theorem @xcite can be faithfully applied for all but a single loop of length eight to determine the flux configuration of the ground state .
an example of the relevant mirror planes to establish this result is shown in fig .
[ fig : lieb ] ( c ) . using numerical calculations
, we verified that also the remaining loop operator has eigenvalue @xmath35 in the ground state .
[ [ projective - symmetries.-2 ] ] projective symmetries. + + + + + + + + + + + + + + + + + + + + + + + sublattice symmetry and , consequently , time - reversal symmetry are implemented trivially for the majorana fermions , i.e. , @xmath78 .
however , when implementing inversion symmetry , we need to supplement it with a gauge transformation that ( artificially ) enlarges the unit cell in the @xmath167-direction . thus , inversion symmetry involves a translation in momentum space by @xmath168 .
( gray lines ) and @xmath169 ( colored lines ) , respectively . ]
[ [ phase - diagram . ] ] phase diagram. + + + + + + + + + + + + + + + the ( 8,3)n lattice is the only one of the lattices considered in this paper that does not exhibit a gapless phase in its phase diagram as shown in fig . [
fig : phase_diagram_8_3n](b ) . instead there are two gapped phases , with fig .
[ fig : energydispersion8n ] showing an example of a gapped dispersion relation for the isotropic coupling point @xmath170 .
the two gapped phases are separated by a line of phase transitions , at which the dispersion exhibits a three - dimensional _ dirac cone _ structure of two doubly degenerate bands ( see fig .
[ fig : energydispersion8nii ] ) .
one way of thinking of such three - dimensional dirac cones is that they are in fact a combination of two ( oppositely charged ) weyl cones of singly degenerate bands .
one way of splitting these weyl cones is by breaking time - reversal symmetry .
indeed , we find that upon applying a magnetic field along the 111 direction the dirac cones split into two weyl nodes each as illustrated in fig . [
fig : energydispersion8nii ] for a small coupling strength @xmath169 . as a result ,
an extended gapless phase emerges in the phase diagram ( around the original line of phase transitions ) as illustrated in the phase diagram of fig .
[ fig : phase_diagram_8_3n ] ( c ) .
this gapless phase ( for small values of @xmath143 ) is thus a weyl spin liquid as discussed already in the context of lattices ( 8,3)b and ( 8,3)c .
one should , however , note that no projective symmetry protects this weyl spin liquid ( as it is the case for the other lattices ) .
in particular , due to the non - trivial implementation of inversion symmetry , the weyl nodes are not generically fixed to zero energy . while for small time - reversal breaking strength @xmath143 they are found to remain strictly at zero energy and move exclusively along high - symmetry lines in the brillouin zone ,
this is no longer true for large values of @xmath143 , where the system eventually develops majorana fermi surfaces .
note that the emerging fermi surfaces are related by a perfect nesting vector @xmath151 as long as inversion symmetry remains intact .
we now turn to the ( 9,3)a lattice , which stands out in our family of tricoordinated , three - dimensional lattices of table [ tab : lattice_overview ] as the only lattice with an _ odd _ number of bonds in the elementary loops .
this oddness has important consequences in particular with regard to the projective time - reversal symmetry of the kitaev model . while the original spin model is time - reversal symmetric , this is no longer the case for the effective majorana model , i.e. , the effective majorana model breaks time - reversal symmetry _
spontaneously_. this has direct consequences for the emergent majorana metal as we will discuss in the following .
[ [ lattice - structure.-4 ] ] lattice structure. + + + + + + + + + + + + + + + + + + + the ( 9,3)a lattice is one of the more complicated lattices with 12 sites per unit cell . in order to simplify notation , we denote all vectors in terms of @xmath171 the complicated value for @xmath172 is needed in order to obtain a lattice with approximately 120 degree bond angles and equal length bonds .
the site positions in the unit cell can be written as @xmath173 with @xmath174 , @xmath175 .
the translation vectors are given by @xmath176 the unit cell and translation vectors are shown in fig .
[ fig : comb_9_a](a ) , as is our choice for assigning the @xmath58- , @xmath59- , and @xmath57-type bonds . up to permutations ,
the bond assignment shown in fig .
[ fig : comb_9_a](a ) is unique when preserving all the lattice symmetries . while @xmath58 and @xmath59 bonds are related by mirror symmetries ,
the @xmath57 bonds are special .
this implies that the physics remains unchanged for interchanging @xmath177 , and the phase diagram is symmetric in @xmath178 and @xmath149 . .
] . ]
we note that an equivalent , though deformed version of the ( 9,3)a lattice can be constructed by joining layers of honeycomb lattices via mid - bond sites , as shown in fig .
[ fig : layered_9_a ] . in our subsequent discussion of the gauge structure
, we will refer to this deformed lattice structure as it is somewhat easier to visualize than the undeformed one .
[ [ gauge - structure - projective - symmetries - and - majorana - metal . ] ] gauge structure , projective symmetries , and majorana metal. + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + the ( 9,3)a lattice has loops with an odd number of bonds , which requires that the corresponding plaquette operators must have eigenvalues @xmath179 or @xmath180 .
any such flux assignment breaks time - reversal symmetry , as acting with time - reversal flips the sign of all plaquette operators without changing the energy of the eigenstate .
note that neither the hamiltonian nor the plaquette operators break time - reversal symmetry both commute with @xmath80 .
thus , the effective majorana model breaks time - reversal symmetry spontaneously and all eigenstates come in time - reversal pairs .
a very similar scenario was put forward in the discussion of yao and kivelson @xcite of a chiral spin liquid ground state emerging for a two - dimensional kitaev model on the 3 - 12 - 12 lattice .
the ( 9,3)a lattice has eight 9-loops per unit cell in the following denoted by @xmath181 but only six of them are linearly independent due to the volume constraints shown in appendix [ ssec : supplemental_9_a ] .
in contrast to lattices that only have plaquettes with an even number of bonds , we need to specify a direction when assigning eigenvalues to the loop operators .
we use the convention that the loop is traversed in the mathematically positive direction when viewed from the center of the unit cell , see fig .
[ fig : canfigure ] . note that this convention implies that inversion maps a loop operator @xmath182 with the loop subscripts @xmath183 as defined in appendix [ ssec : supplemental_9_a ] .
the threefold rotation permutes the loop operators @xmath181 with @xmath184 ( @xmath185 ) cyclically , without changing the eigenvalues .
there are in total two flux configurations , pictorially visualized in fig .
[ fig : canfigure ] , that are compatible with all the lattice symmetries , i.e. , threefold rotation around @xmath54 , inversion with respect to the center of the unit cell , as well as lattice translations .
they differ by the flux ( @xmath48 or @xmath28 ) through the 12-loop defined by @xmath186 i.e. , the loop is given by the product of all bonds within the unit cell at position @xmath43 .
it turns out that the @xmath48-flux configuration has the lower energy of the two .
the reader should note , however , that ( at least for the system sizes we could test numerically ) this symmetric flux sector is _ not _ the ground state sector , even though the energy difference decreases with system size @xcite .
instead , flux configurations that break the threefold rotational symmetry and/or inversion symmetry appear to have slightly lower energy . in the following
, we will briefly discuss the properties of the @xmath48-flux sector .
note that we can always stabilize this sector as the ground - state sector by assigning an energy to 12-loops that are combinations of two adjacent 9-loops , similar to what was done in ref .
@xcite for the kitaev model on the square - octagon lattice ( see appendix [ ssec : supplemental_9_a ] for details ) .
the analysis in this sector turns out to be slightly tedious .
even though the flux configuration itself is translation invariant , it requires a z@xmath0 gauge that enlarges the unit cell by a factor in 2 all three lattice directions , resulting in a 96-site unit cell . in this flux sector
, the system has an extended gapless phase around the isotropic point , shown in fig .
[ fig : comb_9_a](b ) , with a varying number of gapless weyl points , depending on the coupling constants .
when increasing one of the coupling constants sufficiently , all the weyl points annihilate and the system becomes gapped . at the isotropic point , we find that several of the weyl points coincide at @xmath187 , such that the zero - mode is eight - fold degenerate .
however , this multiple zero mode is not stable , and splits into several distinct weyl points as soon as the coupling constants are altered .
in addition to the eightfold zero mode at @xmath187 , there are double weyl nodes at positions @xmath188 with charge @xmath189 , as shown in fig .
[ fig : comb_9_a](c ) .
note that the brillouin zone is computed for the enlarged unit cell . breaking time - reversal symmetry
does not change this physics qualitatively , but only moves the ( double ) weyl nodes in the brillouin zone .
we want to emphasize that for general flux configurations that do not break inversion symmetry , the low - energy physics remains qualitatively the same as in the @xmath48-flux sector discussed above .
in particular , there will be an extended gapless phase around the isotropic point , where the zero - energy modes are weyl points . for inversion - symmetry - breaking flux configurations ,
one generically finds majorana fermi surfaces around the isotropic point .
we complete our classification program by discussing the physics of 3d kitaev models for the three lattices with elementary loop length of 10 ( see table [ tab : lattice_overview ] ) .
the kitaev models for lattices ( 10,3)a and ( 10,3)b have already been discussed in the literature , but for the sake of completeness we briefly review these results here . for the ( 10,3)a or hyperoctagon lattice we summarize the results obtained by some of the authors of this paper in ref . , while for the ( 10,3)b or hyperhoneycomb lattice we report on the results of ref . .
[ [ lattice - structure.-5 ] ] lattice structure. + + + + + + + + + + + + + + + + + + + the ( 10,3)a lattice can be viewed as another higher dimensional variant of the square - octagon lattice , where the squares and octagons form counter - rotating spirals to form a three - dimensional lattice as illustrated in fig . [
fig : comb_10_a](a ) .
close inspection of this spiral structure reveals that it breaks inversion symmetry and as such the ( 10,3)a lattice is one of the few chiral lattices in our family of tricoordinated lattices ( see table [ tab : lattice_overview ] ) . more formally , the ( 10,3)a lattice is a body - centered cubic lattice with four sites per unit cell at positions @xmath190 the lattice vectors are given by @xmath191 and their corresponding reciprocal lattice vectors are @xmath192 the 3d kitaev model for this lattice is defined by assigning bond - directional couplings to the bonds as illustrated in fig .
[ fig : comb_10_a](a ) .
note that all bonds are related to each other by lattice symmetries .
in particular , threefold rotation combined with a permutation of the @xmath58 , @xmath59 , and @xmath57 bonds is a symmetry of the hamiltonian .
thus , the phase diagram is symmetric under permutation of the three different coupling constants . [ [ gauge - structure.-4 ] ] gauge structure. + + + + + + + + + + + + + + + + + for the ( 10,3)a lattice there are two linearly independent elementary loop operators per unit cell , both of which have length 10 . in the following
, we will consider the flux sector , for which all the plaquette operators have eigenvalue @xmath36 , as suggested by lieb s theorem @xcite were it to apply .
we have verified numerically that this is indeed the ground - state flux sector .
in addition , it is the unique flux sector that obeys all the lattice symmetries ; note that choosing all plaquettes to have @xmath28 flux is forbidden due to the volume constraints comprising three adjacent plaquette operators ( see appendix [ ssec : supplemental_10_a ] for a more detailed discussion on these volume constraints ) . .
] parametrizing the magnetic field strength in eq . .
] [ [ projective - symmetries.-3 ] ] projective symmetries. + + + + + + + + + + + + + + + + + + + + + + + the ( 10,3)a lattice has the property that the translation vectors @xmath193 and @xmath100 map the two sublattices onto each other . therefore , as discussed in sec .
[ sec : symmetries ] , sublattice symmetry ( and consequently also time - reversal symmetry ) involves a non - vanishing translation in momentum space by @xmath194 , where @xmath195 and @xmath196 are the reciprocal lattice vectors defined above . as the lattice is chiral , the relevant energy relations are given by particle - hole and time - reversal symmetry @xmath197 [ [ majorana - metal.-2 ] ] majorana metal. + + + + + + + + + + + + + + + + the phase diagram for the kitaev model on lattice ( 10,3)a is shown in fig . [
fig : comb_10_a](b ) ; the system exhibits a gapless phase around the isotropic point , where the gapless modes sit on two majorana fermi surfaces which are visualized in fig .
[ fig : comb_10_a](c ) , see fig .
[ fig : energydispersion10a ] .
the surfaces are centered around the corners of the brillouin zone at @xmath198 and @xmath199 .
the darker shaded orange region of the phase diagram denotes the parameter space where these majorana fermi surfaces are topological , i.e. , they enclose a weyl node at finite energy , a scenario which we discuss in further detail in sec .
[ ssec : topfermisurface ] .
the two majorana fermi surfaces can be mapped onto each other by the perfect nesting vector @xmath77 , as can be seen from fig .
[ fig : comb_10_a](c ) .
this has important consequences .
in particular the system is susceptible to a bcs - type spin - peierls instability @xcite driven by interactions between the majorana fermions , which can be induced by additional spin exchanges such as a heisenberg term augmenting the pure kitaev model .
a short discussion on the spin - peierls instability can be found in sec .
[ sec : spin - peierls ] . breaking time - reversal symmetry does not change the nature of the majorana metal , i.e. , the fermi surfaces remain in place when including the @xmath143 term of eq . .
however , they do deform in a non - trivial way with increasing @xmath143 as illustrated in fig .
[ fig:10a_kappa ] . the ( 10,3)b lattice is probably the best known tricoordinated lattice in three spatial dimensions and is now typically referred to as the hyperhoneycomb lattice @xcite . the 3d kitaev model for this lattice
has recently been discussed extensively @xcite in the context of the iridate @xmath19-li@xmath15iro@xmath16
@xcite , for which spin - orbit entangled @xmath18 moments form on the iridium sublattice , which is precisely the ( 10,3)b hyperhoneycomb lattice . the most symmetric form of the ( 10,3)b hyperhoneycomb lattice can best be visualized as parallel @xmath56-zigzag chains along two distinct directions ( 90@xmath21 rotated with respect to each other @xcite ) that are coupled by @xmath57-bonds [ see figs . [ fig : comparison_10_3_b_c](a ) and ( b ) ] .
it is a close cousin of the third lattice with elementary loop length 10 , the ( 10,3)c lattice .
the latter is made up of three parallel @xmath56-zigzag chains ( 120@xmath21 rotated with respect to each other ) that are coupled by @xmath57-bonds [ see figs . [
fig : comparison_10_3_b_c](c ) and ( d ) ] .
[ [ lattice - structure.-6 ] ] lattice structure. + + + + + + + + + + + + + + + + + + + more formally , the ( 10,3)b lattice is a tetragonal lattice with four sites per unit cell at positions @xmath200 the lattice vectors are given by @xmath201 and their corresponding reciprocal lattice vectors are @xmath202 the unit cell with translation vectors , as well as the assignment of bond types that defines the kitaev hamiltonian are illustrated in fig .
[ fig : phase_diagram_10_3b ] ( a ) .
note that only the @xmath58- and @xmath59-bonds are related by lattice symmetries , which results in a phase diagram that is symmetric under exchange of couplings @xmath148 and @xmath149 .
( parametrizing the effective magnetic field ) along certain high - symmetry lines indicated in the brillouin zone on the right - hand side .
the gray hexagon indicates the plane @xmath203 on which the line of gapless mode ( black line ) is located .
figure adapted from ref . .
] [ [ gauge - structure.-5 ] ] gauge structure. + + + + + + + + + + + + + + + + + for the ( 10,3)b lattice there are two linearly independent loop operators per unit cell , both of which have length 10 ( see appendix [ ssec : supplemental_10_b ] ) . in the following ,
we consider the flux sector for which all loop operators have eigenvalue @xmath36 .
it has been verified numerically in ref .
@xcite , and independently by us , that this is indeed the ground state flux sector .
the vison gap , calculated as the energy gap arising from flipping either an @xmath58- or a @xmath59-bond , is among the highest found for the lattices considered here with @xmath204 in units of the kitaev coupling at the isotropic coupling point @xmath170 . flipping an @xmath58- or a @xmath59-bond operator implies switching signs of six plaquette operators ( see appendix [ ssec : supplemental_10_b ] for a visualization ) .
[ [ projective - symmetries.-4 ] ] projective symmetries. + + + + + + + + + + + + + + + + + + + + + + + as the translation symmetry of this lattice is equivalent to that of its two sublattices , sublattice symmetry and time - reversal symmetry are implemented trivially , i.e. , with vanishing @xmath77 .
as this lattice also possesses inversion symmetry with vanishing @xmath151 , the relevant energy relations become @xmath205 the first originating from particle - hole symmetry and the second from time - reversal or inversion symmetry . [
[ majorana - metal.-3 ] ] majorana metal. + + + + + + + + + + + + + + + + the phase diagram for the kitaev model on lattice ( 10,3)b is shown in fig . [
fig : phase_diagram_10_3b ] ( b ) ; the system exhibits a gapless phase around the isotropic point where the gapless modes form a closed line of dirac nodes , pictured in fig .
[ fig : phase_diagram_10_3b ] ( c ) .
the majorana fermi line , which lies in the @xmath206 plane , is protected by time - reversal symmetry .
breaking time - reversal symmetry causes the fermi line to gap out almost entirely , leaving just two weyl nodes , which are fixed to zero energy as long as inversion symmetry remains intact @xcite ( see fig .
[ fig : energydispersion10b ] ) .
note that the behavior of ( 10,3)b is completely analogous to that of the ( 8,3)c lattice in sec . [ ssec:8c ] , although the details differ .
in particular , we can use the same argument for why time - reversal symmetry breaking can not gap the system completely , and weyl points have to occur . for small values of @xmath143 ,
the weyl points move along the @xmath207-axis up until @xmath208 , where four additional weyl nodes appear .
the full evolution of the weyl nodes and their corresponding fermi arc surface states are visualized in figs .
[ fig:10bfermiarcs ] ( a ) and ( b ) . while the weyl points which move along the @xmath157 axis recombine at @xmath209 , the ones on the front / back surface do not . instead
, one of the velocities of the weyl points vanishes and they become part of gapless nodal lines at @xmath210 .
the ( 10,3)c lattice is a close cousin of ( 10,3)b as already mentioned above . one main distinction between the two lattices is that ( 10,3)c is a chiral lattice , while ( 10,3)b is inversion symmetric .
the chirality has important consequences for the behavior for broken time - reversal , as we will see in the following .
[ [ lattice - structure.-7 ] ] lattice structure. + + + + + + + + + + + + + + + + + + + more formally , the ( 10,3)c lattice is a trigonal lattice with six sites per unit cell at positions @xmath211 the lattice vectors are given by @xmath212 and their corresponding reciprocal lattice vectors are @xmath213 the choice of bond types for implementing the kitaev hamiltonian is illustrated in fig .
[ fig : phase_diagram_10_3c](a ) . note that we chose the @xmath58- and @xmath59-bonds on each of the chains such that the lattice is invariant under a 120@xmath21 screw - rotation .
this ensures that the phase diagram is symmetric under exchanging @xmath177 .
[ [ gauge - structure.-6 ] ] gauge structure. + + + + + + + + + + + + + + + + + for this lattice there are three loop operators of length 10 and three of length 12 per unit cell .
these six loop operators form three closed volumes which leads to only three linearly independent loop operators per unit cell ( see appendix [ ssec : supplemental_10_c ] ) . in what follows
, we consider the flux sector where all loop operators of length 10 have eigenvalue @xmath36 and all loop operators of length 12 have eigenvalue @xmath35 .
this configuration of fluxes respects all lattice symmetries and , although this lattice does not possess the symmetries required for rigorous application of lieb s theorem @xcite , is consistent with the flux assignments one would expect were lieb s theorem to hold . it should be noted that , although this flux configuration breaks no lattice symmetries , fixing a compatible gauge requires an enlargement of the unit cell in the 010-direction ( see appendix [ ssec : supplemental_10_c ] ) .
[ [ projective - symmetries.-5 ] ] projective symmetries. + + + + + + + + + + + + + + + + + + + + + + + as the translation symmetry of this lattice is equivalent to that of its two sublattices , sublattice symmetry and time - reversal symmetry are implemented trivially , i.e. , @xmath78 .
the lattice lacks inversion symmetry , as discussed above , and the relevant energy relations are therefore given by particle - hole and time - reversal symmetry @xmath214 [ [ majorana - metal.-4 ] ] majorana metal. + + + + + + + + + + + + + + + + the phase diagram for the kitaev model for lattice ( 10,3)c is shown in fig . [
fig : phase_diagram_10_3c ] ( b ) .
the model exhibits a gapless phase around the isotropic point where the gapless modes form two closed _ nodal lines _ with a linear dirac - type dispersion in the two directions orthogonal to the nodal line ( see the energy dispersion in fig . [
fig : energydispersion_10_c ] ) . these nodal lines , lying in the @xmath215 plane ,
are fixed to zero energy by time - reversal symmetry .
breaking time - reversal symmetry lifts the dispersion along the original nodal line from zero energy .
in particular , 12 weyl nodes form in the dispersion , whose locations are related by the 120@xmath21 screw rotation . as the ( 10,3)c lattice lacks inversion symmetry ,
the weyl nodes are not fixed to zero energy and are found to indeed move to energies above and below zero energy for arbitrarily small values of @xmath143 . as a result ,
the remaining nodal structure upon breaking of time - reversal symmetry is given by 12 fermi pockets , i.e. , small fermi surfaces that each enclose one of the 12 weyl nodes .
the formation of these fermi surfaces can nicely be tracked in the energy dispersion along high - symmetry lines shown in fig .
[ fig : energydispersion_10_c_kappa ] , where the gray bands show the energy dispersion for the time - reversal invariant system ( exhibiting nodal lines ) and the colored lines correspond to the energy dispersion for @xmath216 ( in units of the kitaev coupling ) .
it can clearly be seen that there are two band crossings between points @xmath39 and @xmath94 ; the one nearer to @xmath39 is located below zero energy and the other above zero energy .
note that because the fermi surfaces enclose a weyl node , they inherit topologically non - trivial features as discussed in section [ ssec : topfermisurface ] in more detail .
finally , we point out that the ( 10,3)c lattice is special in that it is the only lattice for which breaking of time - reversal symmetry _ increases _ the nodal manifold of the majorana metal ( from a line to a surface ) and , thus , also the associated density of states .
a generic feature found in the dispersion relation of _ all _ 3d kitaev models studied in this paper is the occurrence of weyl nodes either right at the fermi energy or above / below it . if the weyl nodes sit right at the fermi energy ( i.e. , zero energy ) , we encounter a spin liquid analog of the electronic weyl semimetal @xcite , a state which we have dubbed a weyl spin liquid in previous work @xcite . if the weyl nodes sit above / below the fermi energy , the system exhibits topological fermi surfaces ( each enclosing at least one weyl node ) , the spin analog of the so - called weyl metal @xcite .
kitaev models in the first category are those defined for lattices ( 8,3)b , ( 8,3)c , ( 8,3)n , ( 9,3)a , and ( 10,3)b , while the kitaev models for lattices ( 8,3)a , ( 10,3)a , and ( 10,3)c are in the second category 9see also table [ tab : majorana_metals ] ) .
we discuss these two scenarios in further detail in the following two subsections .
we will first concentrate on the case where the weyl nodes for one of our 3d kitaev models sit precisely at the fermi energy .
the majorana energy dispersion relation for these systems is then in precise analogy to those of electronic weyl semimetals @xcite .
these electronic weyl semimetals have garnered considerable attention for their recent observation in taas @xcite and photonic materials @xcite .
an intense experimental effort is currently underway to observe the unusual response of these electronic weyl semimetals to electromagnetic fields such as the chiral anomaly @xcite and unusual negative magnetotransport @xcite . for electronic systems
, it has been realized early on that weyl physics can emerge by either breaking time - reversal or inversion symmetry @xcite . for the spin systems at hand it turns out that different symmetry scenarios are at play that give rise to weyl physics .
primarily , we distinguish three different scenarios with respect to the role that time - reversal symmetry plays . in analogy to the electronic systems
, we can find weyl physics when breaking time - reversal symmetry _
explicitly_. this is the case , for instance , for the 3d kitaev models for lattices ( 10,3)b and ( 10,3)c exhibiting nodal lines in the presence of time - reversal symmetry and a number of weyl nodes when breaking time - reversal symmetry .
this also includes the case of the kitaev model for the lattice ( 8,3)n , for which the emergence of weyl nodes arises from the splitting of a dirac cone upon time - reversal symmetry breaking .
this scenario should be carefully distinguished from the physics that plays out for the kitaev model on the lattice ( 9,3)a where time - reversal symmetry is broken _ spontaneously _ and the system is then found to exhibit weyl physics .
the third and most unusual scenario is the one found for the kitaev model for lattice ( 8,3)b . here , _ neither _ time - reversal nor inversion symmetry are broken , but the system nevertheless exhibits weyl physics , a symmetry scenario that is not possible for electronic systems .
the reason that we can observe weyl physics in the spin system without breaking time - reversal nor inversion symmetry is that the _ projective _ time - reversal and inversion symmetry for the underlying majorana fermions are implemented in a non - trivial way .
in particular , these projective symmetries incorporate momentum shifts by @xmath77 and a pair of weyl nodes at @xmath217 always has a pair of time - reversal partners at @xmath218 ( as long as time - reversal symmetry is intact ) .
second , we note that due to particle - hole symmetry ( inherent on the level of the majorana fermions ) , each weyl point at position @xmath123 ( and energy @xmath219 ) will have a particle - hole partner of opposite chirality at position @xmath220 ( and energy @xmath221 ) .
in contrast to the weyl points in an electronic weyl semimetal , one can not regard these as two independent weyl points .
instead , it is more instructive to think of them as a single weyl point of a _ complex _ fermion ( rather than two weyl points of majorana fermions ) .
this is a direct consequence of the fact that the system at hand does not possess @xmath1 symmetry , but only z@xmath0 symmetry .
thus , if neither time - reversal nor inversion symmetry are broken , we will encounter multiples of _ four _ weyl nodes for the underlying majorana system as it is indeed the case for the kitaev model of lattice ( 8,3)b .
note that while each pair of weyl nodes has a corresponding chiral surface state , their effects cancel out each other exactly and the resulting spin liquid is not chiral . upon breaking of time - reversal symmetry ( e.g. by applying a magnetic field ) the perfect nesting between the two pairs of weyl nodes as well as the exact cancellation of their respective fermi arcs are destroyed .
the lack of @xmath1 symmetry in weyl spin liquids has important consequences for various physical observables .
most importantly , weyl spin liquids will not exhibit the usual chiral anomaly as charge pumping between weyl nodes is intimately connected to charge conservation and thus @xmath1 symmetry . instead ,
weyl spin liquids will exhibit a more subtle incarnation of the chiral anomaly , to be discussed elsewhere . on a more formal level
, the different roles of symmetries in the weyl physics of electronic and spin systems is reflected in the classification of their underlying free - fermion hamiltonians in the 10-fold way symmetry classification scheme of altland and zirnbauer @xcite . for electronic systems , weyl semimetals
are found in symmetry classes a or aii , corresponding to the breaking of either time - reversal or inversion symmetry , respectively .
in contrast , for the kitaev models at hand , we find weyl physics in symmetry classes d or bdi depending on whether time - reversal symmetry is broken ( explicitly or spontaneously ) or not .
this distinction of weyl physics in electronic and spin systems will likely impact their sensitivity to disorder , a direction that we will pursue in the future .
we now want to turn to majorana metals where the weyl points occur at _ finite _ energy , which necessarily implies that they are encapsulated by a fermi surface .
such fermi surfaces are called topological fermi surfaces in the following , as they are found to inherit some of the topological features of the enclosed weyl points .
the physical properties discussed here are relevant for the lattices ( 10,3)a and ( 8,3)a , as well as ( 10,3)c for broken time - reversal symmetry .
let us first note that the chern number of a closed 2d surface is still well - defined , as long as it does not cut through any of the fermi surfaces . further the value of the associated chern number can not depend on whether or not the enclosed weyl point sits at zero energy .
thus , even in the presence of fermi surfaces , we find that the effective two - dimensional hamiltonian defined on an arbitrary closed surface that surrounds a topological majorana fermi surface is that of a topological chern insulator . using the same arguments as above
, we find that in the presence of boundaries the majorana fermi surfaces can not lie isolated in the surface brillouin zone , but must again be connected by ( chiral ) fermi arcs , see fig .
[ fig:10_3_a_fermi_arc ] for an example . coupling constant from 0 to @xmath222 where the system becomes gapped .
red ( yellow ) dots denote the position of weyl points of positive ( negative ) chirality .
the black dots are pairs of weyl points of opposite chirality . for @xmath223
the weyl points start to split and annihilate at @xmath224 . ]
the fact that weyl nodes can not be gapped out individually readily implies that the same is true for topological fermi surfaces .
in fact , in order to gap them , one usually has to first annihilate their enclosed weyl nodes pairwise . for the lattices ( 10,3)a and ( 8,3)a
, such pair annihilation can be observed when varying the relative strength of the three kitaev couplings . in particular
, we find that pair annihilation occurs at touching points of two fermi surfaces .
these touching points are , in fact , a pair of weyl nodes of opposite chirality . figures [ fig:10_3_a_jz ] and [ fig:8_3_a_jz ] show the evolution of the fermi surfaces along the line @xmath225 with the position of the enclosed weyl nodes denoted by the red ( positive ) and yellow ( negative ) sphere . for the lattice ( 10,3)a
, the weyl nodes have charge @xmath226 and are located at @xmath227 for @xmath228 .
at @xmath229 they split and pair - annihilate with the weyl nodes of opposite charge at @xmath224 .
the resulting trivial fermi surface persists until @xmath222 , where the system becomes gapped .
the behavior for the kiteav model on lattice ( 8,3)a is slightly different .
the fermi surfaces are topological for @xmath228 .
the touching point at @xmath229 marks the pair creation / annihilation of opposite - charge weyl nodes that split for @xmath223 and annihilate with the original weyl points at @xmath230 , see the sequence of fig .
[ fig:8_3_a_jz ] . .
red ( yellow ) dots denote the position of weyl points of positive ( negative ) chirality .
the black dots are pairs of weyl points of opposite chirality . around @xmath231
each weyl point splits into three . at the isotropic point , two pairs of weyl points with opposite charges are annihilated while one is created at the touching points of the fermi surfaces .
this results in topologically trivial fermi surfaces for @xmath232 . ]
while our entire discussion so far has concentrated on the pure kitaev model , it is of course interesting to also discuss the effect of additional interactions ( e.g. , heisenberg exchange ) on the nature of the gapless spin - liquid ground state .
such additional terms have in general two effects : the vison excitations of the z@xmath0 gauge field gain dynamics and obtain a dispersion , while the majorana fermions generically become interacting .
the former effect may be ignored for sufficiently small perturbations , as the vison excitations remain gapped .
the effect of interactions between majorana fermions depends crucially on the nature of the gapless modes . for quantum spin liquids with a fermi line or weyl points ,
a scaling analysis as , e.g. , done in ref .
@xcite shows that interaction terms are irrelevant at the kitaev point and that the spin liquid is therefore stable against small perturbations . for quantum spin liquids with a majorana fermi surface ,
interactions turn out to be marginal . a careful analysis for the ( 10,3)a hyperoctagon lattice in ref .
@xcite has shown that time - reversal symmetric interactions generically destabilize the majorana fermi surfaces even for infinitesimal coupling strength , and the surfaces gap out except for an odd number of nodal lines . in the following
, we will briefly review the underlying mechanism and the main features of this instability , which will be referred to as a spin - peierls bcs instability for reasons that will become clear in the following .
for further details we refer the reader to ref .
@xcite . as we have seen in section [ sec : symmetries ] , a majorana fermi surface can only be stable if time - reversal symmetry is implemented non - trivially , i.e. , with @xmath233 . combining this relation with particle - hole symmetry , we find that the spectrum necessarily exhibits perfect nesting of the majorana fermi surfaces , @xmath234 .
this perfect nesting for lattices ( 10,3)a and ( 8,3)a is visualized on the left - hand - sides of figs .
[ fig : spinpeierls ] ( a ) and ( b ) , respectively , with the perfect nesting vector indicated by the arrow .
alternatively , we could have chosen to express the system in terms of complex fermions @xmath235 where @xmath236 denote the eigenmodes of the majorana hamiltoninan with @xmath237 for @xmath238 and @xmath239 due to particle - hole symmetry .
the 2@xmath122 majorana fermi surfaces , thus , combine into @xmath122 complex fermi surfaces , and the perfect nesting condition becomes the usual bcs pairing condition @xmath240 , centered around @xmath241 .
note that there is no @xmath1 symmetry in the system .
instead , a non - vanishing pair correlator @xmath242 breaks translation symmetry spontaneously .
the resulting dimerization of the system is reflected , e.g. , in the spin - spin correlations that acquire a staggered component .
due to the spontaneous breaking of translation symmetry , this bcs - type instability shows similar features to the usual spin - peierls instability , except that the dimerization always sets in for infinitesimal interaction strength . as was shown in ref .
@xcite , any additional time - reversal symmetric interaction will , independent of the microscopic details , give rise to this kind of instability . for the kitaev model on lattice
( 10,3)a , time - reversal symmetry ensures that the fermi surface can not be gapped out completely , and an odd number of nodal lines remains . for the kitaev model on lattice ( 8,3)a ,
one finds four majorana surfaces instead of ( the minimal number of ) two . this additional freedom , in principle , allows for interactions to gap out the system completely . ) on the right , respectively .
breaking time - reversal symmetry destroys the perfect nesting condition with wave vector @xmath77 , marked by the arrow . ]
one way to stabilize the majorana fermi surfaces is by breaking time - reversal symmetry .
this leads to a deformation of the majorana fermi surfaces ( as illustrated , e.g. , in fig .
[ fig:10a_kappa ] ) that destroys the perfect nesting condition ,
i.e. , the original nesting vector @xmath77 does not map the two majorana fermi surfaces onto each other any longer .
for the lattice ( 10,3)a , breaking time - reversal symmetry causes elongating / flattening of the surface along the four 111 directions as shown in fig .
[ fig : spinpeierls ] ( a ) ; for ( 8,3)a the surfaces are elongated / shortened along the 001 direction as shown in fig .
[ fig : spinpeierls ] ( b ) .
in both cases , translation along @xmath77 does not map the surfaces exactly onto each other .
the resulting mismatch in energy cuts off the bcs instability at low enough temperatures and restores the fermi surface .
the physics of fractionalization , accompanied by the formation of spin - liquid ground states , is beautifully embodied in the kitaev model .
the model is unique in that it allows to precisely track on an analytical level the splitting of the original spin degrees of freedom into majorana fermions and a z@xmath0 gauge field .
this study has explored this phenomenon for three - dimensional kitaev models with a focus on the collective physics of the itinerant majorana fermions , the formation of majorana metals whose nature intimately depends on the topology of the underlying lattice .
we have provided a comprehensive classification of these majorana metals for the most elementary tricoordinated lattices in three spatial dimensions ( summarized in tables [ tab : lattice_overview ] and [ tab : majorana_metals ] ) , which is rooted in an elementary symmetry analysis of the projective time - reversal and inversion symmetries for these lattices .
focusing primarily on the majorana physics , our study already attests to the rich physics of three - dimensional kitaev models , while also pointing to a number of future research directions which we briefly comment on in the following . on a conceptual level
, it will be interesting to complement the current analysis with a more rigorous study of the physics of the z@xmath0 gauge field for the family of three - dimensional kitaev models at hand .
even for the pure kitaev model , for which the gauge field remains static , the ground state of the gauge field might be somewhat non - trivial for some of the lattices under scrutiny in this study .
in particular , for lattice ( 9,3)a we have found possible evidence that some of the low - energy states or even the ground state of the z@xmath0 gauge field might break some of the point - group symmetries of the lattice . to further elucidate this possibility , a more stringent numerical approach is needed such as the monte carlo sampling approach recently developed by the motome group @xcite .
such an unbiased , ergodic sampling approach would also be helpful in understanding the low - energy physics of the z@xmath0 gauge field for lattice ( 8,3)c , for which we find evidence of geometric frustration in the assignment of z@xmath0 fluxes , possibly leading to an extensive degeneracy of gauge field configurations in the ground state
. it will be interesting to further explore how this possibly somewhat unusual physics of the z@xmath0 gauge field couples back to the formation of a collective state of the majorana fermions .
one key distinction between kitaev models in two and three spatial dimensions is their finite - temperature behavior .
for the three - dimensional kitaev models we expect to observe a finite - temperature phase transition at which the z@xmath0 gauge field orders @xcite .
indeed such a finite - temperature phase transition has recently been observed in monte carlo simulations of the ( 10,3)b hyperhoneycomb lattice @xcite .
generically , this transition is expected to be a continuous phase transition in the inverted 3d ising universality class . however
, this must not be the case for lattice geometries where the zero - temperature physics indicates a breaking of time - reversal and/or point - group symmetries in addition to the z@xmath0 symmetry associated with the gauge theory
. it will be interesting to numerically explore whether these systems exhibit a _
single _ continuous finite - temperature phase transition , at which simultaneously multiple symmetries are broken spontaneously . if so , they will establish remarkable instances of lattice models whose critical behavior evades a description in terms of the landau - ginzburg - wilson ( lgw ) paradigm .
in fact , such unconventional finite - temperature transitions in kitaev models might provide the most `` natural '' instances of non - lgw criticality in three - dimensional systems , which so far has been explored in the context of classical dimer @xcite and loop @xcite models somewhat more artificially constructed systems that have no direct realization in a microscopic setting .
complementary to this study of the gapless spin liquids arising for roughly equal coupling strength @xmath5 in three - dimensional kitaev models , it should be compelling to systematically investigate the gapped spin liquids which arise when one of the three couplings dominates over the other two .
in particular , it will be interesting to go beyond initial studies of the ( 10,3)b hyperhoneycomb lattice @xcite and see whether some of these gapped phases give rise to non - trivial loop statistics @xcite .
another important avenue will be to study the effect of disorder in these three - dimensional kitaev models .
such an analysis is of particular interest for the weyl spin liquids found in a number of lattice geometries ( see table [ tab : majorana_metals ] and the discussion in sec . [
sec : weyl ] ) . what sets these weyl spin liquids apart from their electronic counterparts is their respective symmetry classification in terms of the ten - fold way classification scheme of free - fermion systems @xcite .
weyl semimetals in electronic systems can arise from breaking either time - reversal or inversion symmetry , putting these systems into symmetry classes a or aii , respectively .
in contrast , weyl physics in majorana systems ( which per se exhibit particle - hole symmetry ) can arise from breaking time - reversal symmetry ( akin to the electronic system ) but also without breaking of time - reversal symmetry _ nor _ inversion symmetry , putting the majorana systems into symmetry classes d or bdi , respectively .
it will be interesting to identify to what extent disorder physics in weyl semimetals is sensitive to the underlying symmetry class .
one of the most salient future research directions will be to go beyond the pure kitaev model and to study the physics arising from additional spin exchanges such as a heisenberg exchange .
these magnetic interactions have two effects .
they render dynamics to the z@xmath0 gauge field , allowing the elementary vison excitations to disperse through the lattice . at the same time
these additional spin exchanges also induce interactions between the majorana fermions , possibly destabilizing the majorana metal of the pure kitaev model as discussed in sec .
[ sec : spin - peierls ] on the spin - peierls instability of majorana metals with a majorana fermi surface @xcite .
finally , we hope that our study highlighting the rich physics arising from fractionalization in three - dimensional kitaev models will provide further stimulus to the ongoing search for materials realizing spin - orbit entangled @xmath18 mott insulators with strong bond - directional kitaev - type exchanges
. it would be most enthralling if candidate @xmath18 mott materials for tricoordinated lattice structures beyond the ( 10,3)b hyperhoneycomb lattice of @xmath19-li@xmath15iro@xmath16 ( and its higher harmonic in @xmath20-li@xmath15iro@xmath16 ) could be realized .
while certainly not the focus of the current study , we might offer some minimal guidance for this experimental search in noting that one prerequisite facilitating the emergence of strong bond - directional kitaev exchanges is the occurrence of double exchange paths between the spin - orbit entangled moments on the tricoordinated lattice structure @xcite .
one way to realize such a scenario is found in the iridates @xmath243iro@xmath244 ( with @xmath245na , li ) where @xmath246 ir@xmath247 ions are embedded in iro@xmath248 oxygen cages that form an _ edge - sharing _ network .
such an edge - sharing network structure can also be realized for the lattices ( 10,3)a and ( 8,3)b visualized in fig .
[ fig : materials ] in addition to the experimentally observed realization of the lattice ( 10,3)b in @xmath19-li@xmath15iro@xmath16
. it would be thus highly tantalizing if spin - orbit entangled @xmath18 mott insulators could be realized for one of these lattice structures .
we thank l. balents , y. motome , j. nasu , a. rosch , and a. vishwanath for insightful discussions .
m.h . acknowledges partial support through the emmy - noether program of the dfg .
the numerical simulations were performed on the cheops cluster at rrzk cologne .
all authors acknowledge hospitality of the kitp during the final stages of this work .
this research was supported in part by the national science foundation under grant no .
nsf phy11 - 25915 .
00 f. d. m. haldane , ` luttinger liquid theory ' of one - dimensional quantum fluids .
i. properties of the luttinger model and their extension to the general 1d interacting spinless fermi gas , http://dx.doi.org/10.1088/0022-3719/14/19/010[j .
c * 14 * , 2585 ( 1981 ) ] ; effective harmonic - fluid approach to low - energy properties of one - dimensional quantum fluids , http://dx.doi.org/10.1103/physrevlett.47.1840[phys .
rev . lett .
* 47 * , 1840 ( 1981 ) ] ; j. voit , one - dimensional fermi liquids , http://dx.doi.org/10.1088/0034-4885/58/9/002[rep . prog .
phys . * 58 * , 977 ( 1995 ) ] ; o. auslaender , h. steinberg , a. yacoby , y. tserkovnyak , b. i. halperin , k. w. baldwin , l. n. pfeiffer , and k. w. west , spin - charge separation and localization in one - dimension , link:\doibase 10.1126/science.1107821[science * 308 * , 88 ( 2005 ) ] .
d. c. tsui , h. l. stormer , and a. c. gossard , two - dimensional magnetotransport in the extreme quantum limit , http://dx.doi.org/10.1103/physrevlett.48.1559[phys .
* 48 * , 1559 ( 1982 ) ] . c. castelnovo , r. moessner , and s.l .
sondhi , magnetic monopoles in spin ice , link:\doibase 10.1038/nature06433[nature ( london ) * 451 * , 42 ( 2008 ) ] .
p. milde , d. khler , j. seidel , l. m. eng , a. bauer , a. chacon , j. kindervater , s. mhlbauer , c. pfleiderer , s. buhrandt , c. schtte , and a. rosch , unwinding of a skyrmion lattice by magnetic monopoles , link:\doibase 10.1126/science.1234657[science * 340 * ,
1076 ( 2013 ) ] . for a recent review see e.g. l. balents , spin liquids in frustrated magnets , link:\doibase 10.1038/nature08917[nature * 464 * , 199 ( 2010 ) ] .
n. read and s. sachdev , large-@xmath249 expansion for frustrated quantum antiferromagnets , http://dx.doi.org/10.1103/physrevlett.66.1773[phys .
rev . lett . * 66 * , 1773 ( 1991 ) ] .
t. senthil and m. p. a. fisher , gauge theory of electron fractionalization in strongly correlated systems , http://dx.doi.org/10.1103/physrevb.62.7850[phys .
b * 62 * , 7850 ( 2000 ) ] . x .- g .
wen , quantum orders and symmetric spin liquids , http://dx.doi.org/10.1103/physrevb.65.165113[phys .
b * 65 * , 165113 ( 2002 ) ] .
a. kitaev , anyons in an exactly solved model and beyond , link:\doibase 10.1016/j.aop.2005.10.005[ann . phys .
* 321 * , 2 ( 2006 ) . ]
g. khaliullin , orbital order and fluctuations in mott insulators , link:\doibase 10.1143/ptps.160.155[prog .
. suppl . * 160 * , 155 ( 2005 ) ] ; g. jackeli and g. khaliullin , mott insulators in the strong spin - orbit coupling limit : from heisenberg to a quantum compass and kitaev models , link:\doibase 10.1103/physrevlett.102.017205[phys .
* 102 * , 017205 ( 2009 ) ] ; j. chaloupka , g. jackeli , and g. khaliullin , kitaev - heisenberg model on a honeycomb lattice : possible exotic phases in iridium oxides a@xmath11iro@xmath244 , link:\doibase 10.1103/physrevlett.105.027204[phys .
105 * , 027204 ( 2010 ) ] .
y. singh , s. manni , j. reuther , t. berlijn , r. thomale , w. ku , s. trebst , and p. gegenwart , relevance of the heisenberg - kitaev model for the honeycomb lattice iridates a@xmath11iro@xmath244 , link:\doibase 10.1103/physrevlett.108.127203[phys .
* 108 * , 127203 ( 2012 ) ] ; s. k. choi _
et al . _ ,
spin waves and revised crystal structure of honeycomb iridate na@xmath11iro@xmath244 , link:\doibase 10.1103/physrevlett.108.127204[phys .
lett . * 108 * , 127204 ( 2012 ) ] ; r. comin _
et al . _ , na@xmath11iro@xmath244 as a novel relativistic mott insulator with a 340-mev gap , link:\doibase 10.1103/physrevlett.109.266406[phys .
109 * , 266406 ( 2012 ) ] ; feng ye , songxue chi , huibo cao , bryan c. chakoumakos , jaime a. fernandez - baca , radu custelcean , t. f. qi , o. b. korneta , and g. cao , direct evidence of a zigzag spin - chain structure in the honeycomb lattice : a neutron and x - ray diffraction investigation of single - crystal na@xmath11iro@xmath244 , link:\doibase 10.1103/physrevb.85.180403[phys .
b * 85 * , 180403 ( 2012 ) ] ; h. gretarsson _
_ , crystal - field splitting and correlation effect on the electronic structure of a@xmath11iro@xmath244 , link:\doibase 10.1103/physrevlett.110.076402[phys .
lett . * 110 * , 076402 ( 2013 ) ] ; h. gretarsson _
et al . _ ,
magnetic excitation spectrum of na@xmath11iro@xmath244 probed with resonant inelastic x - ray scattering , link:\doibase 10.1103/physrevb.87.220407[phys .
b * 87 * , 220407 ( 2013 ) ] .
k. w. plumb , j. p. clancy , l. j. sandilands , v. vijay shankar , y. f. hu , k. s. burch , hae - young kee , and young - june kim , @xmath17-rucl@xmath244 : a spin - orbit assisted mott insulator on a honeycomb lattice , http://dx.doi.org/10.1103/physrevb.90.041112[phys . rev .
b * 90 * , 041112(r ) ( 2014 ) ] ; luke j. sandilands , yao tian , k. w. plumb , young - june kim , and kenneth s. burch , scattering continuum and possible fractionalized excitations in @xmath17-rucl@xmath244 http://dx.doi.org/10.1103/physrevlett.114.147201[phys .
lett . * 114 * , 147201 ( 2015 ) ] ; v. vijay shankar , heung - sik kim , and hae - young kee , kitaev magnetism in honeycomb rucl@xmath244 with intermediate spin - orbit coupling , http://dx.doi.org/10.1103/physrevb.91.241110[phys .
b * 91 * , 241110 ( 2015 ) ] ; m. majumder , m. schmidt , h. rosner , a. a. tsirlin , h. yasuoka , and m. baenitz , anisotropic ru3 + 4d5 magnetism in the alpha - rucl3 honeycomb system : susceptibility , specific heat and zero field nmr , http://dx.doi.org/10.1103/physrevb.91.180401[phys . rev .
b * 91 * , 180401(r ) ( 2015 ) ] ; l. j. sandilands , y. tian , a. a. reijnders , h .- s .
kim , k. w. plumb , h .- y .
kee , y .- j .
kim , and k. s. burch , orbital excitations in the 4d spin - orbit coupled mott insulator @xmath17-rucl@xmath244 , http://arxiv.org/abs/1503.07593[arxiv:1503.07593 ] ; y. kubota , h. tanaka , t. ono , y. narumi , and k. kindo , successive magnetic phase transitions in @xmath17-rucl@xmath244 :
xy - like frustrated magnet on the honeycomb lattice http://dx.doi.org/10.1103/physrevb.91.094422[phys .
b * 91 * , 094422 ( 2015 ) ] ; a. banerjee , c.a .
bridges , j - q .
yan , a.a .
aczel , l. li , m.b .
stone , g.e .
granroth , m.d .
lumsden , y. yiu , j. knolle , d.l .
kovrizhin , s. bhattacharjee , r. moessner , d.a .
tennant , d.g .
mandrus , and s.e .
nagler , proximate kitaev quantum spin liquid behaviour in @xmath17-rucl@xmath244 , http://arxiv.org/abs/1504.08037[arxiv:1504.08037 ] .
sae hwan chun , jong - woo kim , jungho kim , h. zheng , constantinos c. stoumpos , c. d. malliakas , j. f. mitchell , kavita mehlawat , yogesh singh , y. choi , t. gog , a. al - zein , m. moretti sala , m. krisch , j. chaloupka , g. jackeli , g. khaliullin , and b. j. kim , direct evidence for dominant bond - directional interactions in a honeycomb lattice iridate na@xmath11iro@xmath244 , link:\doibase 10.1038/nphys3322[nature physics * 11 * , 462 ( 2015 ) ] .
t. takayama , a. kato , r. dinnebier , j. nuss , and h. takagi , hyper - honeycomb iridate @xmath19-li@xmath11iro@xmath244 as a platform for kitaev magnetism , \doibase http://dx.doi.org/10.1103/physrevlett.114.077202[phys .
lett . * 114 * , 077202 ( 2015 ) ] .
k. a. modic , t. e. smidt , i. kimchi , n. p. breznay , a. biffin , s. choi , r. d. johnson , r. coldea , p. watkins- curry , g. t. mccandless , j. y. chan , f. gandara , z. islam , a. vishwanath , a. shekhter , r. d. mcdonald , and j. g. analytis , realization of a three - dimensional spin - anisotropic harmonic honeycomb iridate , link:\doibase 10.1038/ncomms5203[nature comm . * 5 * , 4203 ( 2014 ) ] .
early attempts to construct three - dimensional kitaev models include the work by s. ryu @xcite discussing a spin-3/2 kitaev - type model on the four - coordinated diamond lattice , as well as the work by si and yu @xcite identifying spin-1/2 kitaev models on certain three - dimensional , tricoordinated helix lattices .
s. ryu , three - dimensional topological phase on the diamond lattice , http://dx.doi.org/10.1103/physrevb.79.075124[phys .
b * 79 * , 075124 ( 2009 ) ] .
tieyan si and yue yu , exactly soluble spin-1/2 models on three - dimensional lattices and non - abelian statistics of closed string excitations , http://arxiv.org/abs/0709.1302[arxiv:0709.1302 ] ; tieyan si and yue yu , anyonic loops in three dimensional spin liquid and chiral spin liquid , http://dx.doi.org/10.1016/j.nuclphysb.2008.06.009[nucl .
b * 803 * , 428 ( 2008 ) ] .
m. hermanns and s. trebst , quantum spin liquid with a majorana fermi surface on the three - dimensional hyperoctagon lattice , link:\doibase 10.1103/physrevb.89.235102[phys .
b * 89 * , 235102 ( 2014 ) . ] s. mandal and n. surendran , exactly solvable kitaev model in three dimensions , link:\doibase 10.1103/physrevb.79.024426[phys . rev .
b * 79 * , 024426 ( 2009 ) . ]
m. hermanns , k. obrien , and s. trebst , weyl spin liquids , \doibase http://dx.doi.org/10.1103/physrevlett.114.157202[phys . rev
. lett . * 114 * , 157202 ( 2015 ) ] . for a given pair of nearest - neighbor bonds , the _ elementary loop _
is defined by the smallest loop containing this pair of bonds .
note that the set of elementary loops is not necessarily identical to the set of _ fundamental loops_. the latter denotes the set of ( independent ) loops , from which any loop in the lattice can be built , and may contain longer loops .
a. f. wells , three - dimensional nets and polyhedra , wiley ( 1977 ) .
none of these criteria are essential for the physical properties , which are solely determined by the topology of the lattice . however , these criteria select a subset of lattices that exhibit particularly simple , regular structures and that already showcase all the different spin liquid phases that are allowed by our symmetry analysis .
h. heesch and f. laves , ber dnne kugelpackungen , z. kristallogr .
* 85 * , 443 ( 1933 ) .
t. sunada , crystals that nature might miss creating , http://www.ams.org/notices/200802/tx080200208p.pdf[notices of the ams * 55 * , 208 ( 2008 ) ] . for the lattices ( 9,3)a and ( 8,3)c , the flux sector we consider is not the ground state sector for the pure kitaev model .
however , one can always stabilize it as the ground state sector by adding additional terms to the kitaev hamiltonian that favor / penalize flux through the plaquettes .
for such an augmented hamiltonian , also the lattices ( 9,3)a and ( 8,3)c have fully gapped vison excitations throughout the phase diagram .
a. p. schnyder , s. ryu , a. furusaki , and a. w. w. ludwig , classification of topological insulators and superconductors in three spatial dimensions , http://dx.doi.org/10.1103/physrevb.78.195125 [ phys .
b * 78 * , 195125 ( 2008 ) ] ; a. y. kitaev , periodic table for topological insulators and superconductors , http://dx.doi.org/10.1063/1.3149495[aip conf . proc . * 1134 * , 22 ( 2009 ) ] .
s. ryu , a. p. schnyder , a. furusaki , and a. w. w. ludwig , topological insulators and superconductors : ten - fold way and dimensional hierarchy , http://dx.doi.org/10.1088/1367-2630/12/6/065010[new j. phys .
* 12 * , 065010 ( 2010 ) ] .
a. altland and m. r. zirnbauer , nonstandard symmetry classes in mesoscopic normal - superconducting hybrid structures , http://dx.doi.org/10.1103/physrevb.55.1142[phys .
b * 55 * , 1142 ( 1997 ) ] . for lattices where @xmath78 , we usually choose the directionality from @xmath95 to @xmath250 lattice . for lattices with @xmath251 or the non - bipartite lattice ( 9,3)a
, the reader is referred to appendix [ sec : supplemental ] where we provide the details of our choice of directionality .
lieb , flux phase of the half - filled band , http://dx.doi.org/10.1103/physrevlett.73.2158[phys .
* 73 * , 2158 ( 1994 ) ] .
k. momma and f. izumi , vesta 3 for three - dimensional visualization of crystal , volumetric and morphology data , link:\doibase 10.1107/s0021889811038970[j .
. crystallogr . * 44 * , 1272 ( 2011 ) ] .
see supplemental material at url for vesta files of all the lattices considered here .
s. yang , d. l. zhou , and c. p. sun , mosaic spin models with topological order , http://dx.doi.org/10.1103/physrevb.76.180404[phys .
b * 76 * , 180404(r ) ( 2007 ) . ] h. yao and s. a. kivelson , exact chiral spin liquid with non - abelian anyons , http://dx.doi.org/10.1103/physrevlett.99.247203[phys .
lett . * 99 * , 247203 ( 2007 ) ] .
m. hermanns , s. trebst , and a. rosch , spin - peierls instability of three - dimensional spin liquids with majorana fermi surfaces , http://dx.doi.org/10.1103/physrevlett.115.177205[phys .
lett . * 115 * , 177205 ( 2015 ) ] .
x. wan , a. m. turner , a. vishwanath , and s. y. savrasov , topological semimetal and fermi - arc surface states in the electronic structure of pyrochlore iridates , http://dx.doi.org/10.1103/physrevb.83.205101[phys .
b * 83 * , 205101 ( 2011 ) ] . hsin - hua lai and olexei i. motrunich , su(2)-invariant majorana spin liquid with stable parton fermi surfaces in an exactly solvable model , http://dx.doi.org/10.1103/physrevb.84.085141[phys . rev .
b * 84 * , 085141 ( 2011 ) . ] as the nodal structure of the ( 8,3)c lattice is rather non - generic at the isotropic point , one should consider anisotropic couplings where the lines form contractable loops for this argument . as the unit cell for this flux configuration contains already 96 sites , a thorough finite - size scaling analysis
is extremely hard .
hsin - hua lai and olexei i. motrunich power - law behavior of bond energy correlators in a kitaev - type model with a stable parton fermi surface , http://dx.doi.org/10.1103/physrevb.83.155104[phys .
b * 83 * , 155104 ( 2011 ) . ]
e. k .- h .
lee , r. schaffer , s. bhattacharjee , and y. b. kim , heisenberg - kitaev model on the hyperhoneycomb lattice , http://dx.doi.org/10.1103/physrevb.89.045117[phys .
b * 89 * , 045117 ( 2014 ) ] . i. kimchi , j. g. analytis , and a. vishwanath , three - dimensional quantum spin liquids in models of harmonic - honeycomb iridates and phase diagram in an infinite - d approximation , link:\doibase 10.1103/physrevb.90.205126[phys . rev .
b * 90 * , 205126 ( 2014 ) ] .
j. nasu , t. kaji , k. matsuura , m. udagawa , y. motome , finite - temperature phase transition to a quantum spin liquid in a three - dimensional kitaev model on a hyperhoneycomb lattice , http://dx.doi.org/10.1103/physrevb.89.115125[phys .
b * 89 * , 115125 ( 2014 ) ] . while the most symmetric incarnation of the ( 10,3)b lattice exhibits 90 degree angles between the zig - zag chains ( see fig . [
fig : comparison_10_3_b_c ] ) , we use a smooth deformation of the lattice in our calculations which corresponds to the hyperhoneycomb lattice structure studied in the context of @xmath19-li@xmath11iro@xmath244 @xcite . for a recent review ,
see e.g. a. a. burkov , chiral anomaly and transort in weyl metals , link:\doibase 10.1088/0953 - 8984/27/11/113201[j .
phys . : condens .
matter * 27 * , 113201 ( 2015 ) . ]
b. q. lv et al . , observation of weyl nodes in taas , link:\doibase 10.1038/nphys3426[nature physics * 11 * , 724 ( 2015 ) ] ; su - yang xu et al . ,
discovery of a weyl fermion semimetal and topological fermi arcs , link:\doibase 10.1126/science.aaa9297[science * 349 * , 613 ( 2015 ) ] .
ling lu et al . ,
experimental observation of weyl points , link:\doibase 10.1126/science.aaa9273[science * 349 * , 622 ( 2015 ) ] .
a. a. zyuzin and a. a. burkov , topological response in weyl semimetals and the chiral anomaly http://dx.doi.org/10.1103/physrevb.86.115133[phys .
b * 86 * , 115133 ( 2012 ) ] . jun xiong et al . , evidence for the chiral anomaly in the dirac semimetal na@xmath244bi , link:\doibase 10.1126/science.aac6089[science * 350 * , 413 ( 2015 ) ] .
xiaochun huang et al . ,
observation of the chiral anomaly induced negative magneto - resistance in 3d weyl semi - metal taas , link:10.1103/physrevx.5.031023[phys .
x * 5 * , 031023 ( 2015 ) . ]
j. nasu , m. udagawa , and y. motome , vaporization of kitaev spin liquids , http://dx.doi.org/10.1103/physrevlett.113.197205[phys .
lett . * 113 * , 197205 ( 2014 ) ] .
g. chen , j. gukelberger , s. trebst , f. alet , and l. balents , coulomb gas transitions in three - dimensional classical dimer models , http://dx.doi.org/10.1103/physrevb.80.045112[phys .
b * 80 * , 045112 ( 2009 ) ] .
a. nahum , j. t. chalker , p. serna , m. ortuo , and a. m. somoza , deconfined quantum criticality , scaling violations , and classical loop models , link:\doibase 10.1103/physrevx.5.041048[phys .
x * 5 * , 041048 ( 2015 ) ] c. wang and m. levin , braiding statistics of loop excitations in three dimensions , http://dx.doi.org/10.1103/physrevlett.113.080403[phys .
* 113 * , 080403 ( 2014 ) ] .
in this first appendix we expand the discussion of the lattice structure for each of the tricoordinated lattices of table [ tab : lattice_overview ] .
in particular , we show visualizations of each lattice structure along various high - symmetry projections , with the underlying vesta @xcite visualization files provided in the supplemental material of this paper .
the latter also define the crystallographic axes referred to in the captions .
in addition , we give detailed information on the space group of each lattice and the wyckoff positions for their unit cell .
this lattice is bipartite with @xmath256 , where @xmath257 is a reciprocal lattice vector defined in eq .
, due to @xmath257 connecting different sublattices , and lacks inversion symmetry .
there are two distinct sets of @xmath58- , @xmath59- and @xmath57-bonds which can not be related by symmetries , those which make up the co - rotating spirals [ see fig .
[ fig : comparison_8_3_a_b](a ) ] and those that connect two nearest - neighbor spirals .
all bonds of a given set are related by a combination of @xmath258 symmetry and a three - fold screw rotation .
the symmetry between @xmath58- , @xmath59- and @xmath57-bonds is reflected in a phase diagram symmetric in all couplings . the lattice ( 8,3)a is depicted in certain high - symmetry projections in fig . [
fig : sym_8_3_a ] : in ( a ) the lattice is viewed along a three - fold screw axis and ( b ) shows the lattice along a two - fold rotation axis .
this lattice is bipartite with @xmath264 , where @xmath265 and @xmath257 are reciprocal lattice vectors defined in eq . , due to @xmath265 and @xmath257 connecting different sublattices , and possesses inversion symmetry with vanishing @xmath151 .
there are two distinct sets of @xmath58- , @xmath59- and @xmath57-bonds which can not be related by symmetries , those which make up the counter - rotating spirals [ see fig .
[ fig : comparison_8_3_a_b](b ) ] and those which connect them .
all bonds of a given set are related by a combination of @xmath266 and inversion symmetries . the symmetry between @xmath58- , @xmath59- and @xmath57-bonds is reflected in a phase diagram symmetric in all couplings . the lattice ( 8,3)b is depicted in certain high - symmetry projections in fig . [
fig : sym_8_3_b ] : in ( a ) the lattice is viewed along a three - fold rotation axis and ( b ) shows the lattice along a two - fold rotation axis .
this lattice is bipartite with vanishing @xmath77 and possesses inversion symmetry with vanishing @xmath151 .
there are two distinct sets of @xmath58- , @xmath59- and @xmath57-bonds ; those forming the zig - zag chains along @xmath54 , and those lying in the @xmath56-plane .
the sets of all @xmath58- , @xmath59- and @xmath57-bonds are related to each other by a six - fold screw rotation , which is reflected in a phase diagram symmetric in all couplings .
lattice ( 8,3)n is described by the tetragonal space group @xmath273 ( no .
139 ) with @xmath274 .
the wyckoff positions for the unit cell are @xmath275 @xmath276 with @xmath277 , and @xmath278 @xmath279 with @xmath277 and @xmath280 .
this lattice is bipartite with vanishing @xmath77 and possesses inversion symmetry with @xmath281 , where @xmath257 is a reciprocal lattice vector defined in eq . .
all @xmath58- and @xmath59-bonds are related by a combination of inversion , four - fold rotation , and mirror symmetries .
the @xmath57-bonds come in two distinct sets : those that lie in the @xmath56 plane and connect nearest - neighbor `` squares '' , and those that are along @xmath54 and connect neighboring `` square - octagon planes . '' within each set , the bonds are related to each other by four - fold rotation symmetry .
however , @xmath57-bonds are not related to any other bond type by lattice symmetries .
the symmetry between @xmath58- and @xmath59- bonds is reflected in a phase diagram symmetric under @xmath282 . the lattice ( 8,3)n is depicted in certain high - symmetry projections in fig . [
fig : sym_8_3_n ] in ( a ) the lattice is viewed along a two - fold rotation axis and ( b ) shows the lattice along a four - fold rotation axis . lattice ( 9,3)a is described by the trigonal space group @xmath261 ( no .
166 ) with @xmath285 .
the wyckoff positions for the ( hexagonal ) unit cell are @xmath263 @xmath286 with @xmath287 , @xmath288 , and @xmath289 @xmath290 with @xmath291 and @xmath292 .
this lattice is the only _
non_-bipartite lattice considered in this paper .
it possesses inversion symmetry with vanishing @xmath151 .
all @xmath58- and @xmath59-bonds are related by a combination of @xmath266 and mirror symmetries .
there are two distinct sets of @xmath57-bonds which are not related by any symmetries , however , all bonds of a given set may be mapped to each other via @xmath266 symmetry .
the symmetry between @xmath58- and @xmath59-bonds is reflected in a phase diagram symmetric under @xmath282 .
this lattice is bipartite with @xmath297 , where @xmath298 and @xmath257 are reciprocal lattice vectors defined in eq . , but lacks inversion symmetry .
all @xmath58- , @xmath59- and @xmath57-bonds are related by a combination of @xmath266 symmetry and a four - fold screw rotation , which is reflected in a phase diagram symmetric in all couplings . the lattice ( 10,3)a is depicted in certain high - symmetry projections in fig . [
fig : sym_10_3_a ] : in ( a ) the lattice is viewed along a four - fold screw axis and shows the lattice in its square - octagon projection , ( b ) shows the lattice along a two - fold rotation axis , and ( c ) shows the lattice viewed along a three - fold rotation axis .
this lattice is bipartite with vanishing @xmath77 and possesses inversion symmetry with vanishing @xmath151 .
all @xmath58- and @xmath59-bonds are related by a combination of @xmath258 symmetry and a two - fold screw rotation .
additionally , all @xmath57-bonds are related to each other by inversion symmetry , but are not related to any other bond type by lattice symmetries .
the symmetry between @xmath58- and @xmath59-bonds is reflected in a phase diagram symmetric under @xmath282 . the lattice ( 10,3)b is depicted in certain high - symmetry projections in fig .
[ fig : sym_10_3_b ] in ( a ) the lattice is viewed along the four - fold screw rotation axis and ( b ) shows the lattice along a two - fold screw axis .
this lattice is bipartite with vanishing @xmath77 , but lacks inversion symmetry .
all @xmath58- and @xmath59-bonds are related by a six - fold screw rotation .
additionally , all @xmath57-bonds are related to each other by the same six - fold screw rotation , but are not related to any other bond type by lattice symmetries .
the symmetry between @xmath58- and @xmath59-bonds is reflected in a phase diagram symmetric under @xmath282 . the lattice ( 10,3)c is depicted in certain high - symmetry projections in fig . [
fig : sym_10_3_c ] in ( a ) the lattice is viewed along a three - fold screw axis and shows the lattice in its kagome projection and ( b ) shows the lattice along a two - fold screw axis . in this second appendix
, we provide additional information on the 3d kitaev models for the tricoordinated lattices reviewed in the first appendix .
our focus here is on a detailed expos of the gauge structure of the ground state of the kitaev model defined on one of these lattices .
in particular , we give an explicit expression for the kitaev hamiltonian in its majorana representation in this gauge .
the lattice ( 8,3)a possesses three loop operators of length 8 and three of length 14 per unit cell .
these six loop operators can be combined to form three closed volumes , each of which must have vanishing total flux , resulting in only three linearly independent loop operators per unit cell .
one of these closed volumes is illustrated in fig .
[ fig : volumes_8_3a ] .
the remaining two volumes are related by a three - fold screw rotation .
the smallest vison loop in this lattice threads two plaquettes of length 8 , as well as several plaquettes of length 14 , as visualized in fig .
[ fig : vison_8_3a ] . the calculations for lattice
( 8,3)a were performed using the following reference gauge : @xmath315 in this gauge , the momentum space hamiltonian reads as @xmath316 where @xmath317 the gauge - fixed matrix representations of the symmetry operators relevant to our classification scheme are @xmath318 the lattice ( 8,3)b possesses three loop operators of length 8 and one of length 12 per unit cell .
these four loop operators can be combined with four loop operators from neighboring unit cells to form a closed volume which must have vanishing total flux , resulting in only three linearly independent loop operators per unit cell .
this closed volume is illustrated in fig .
[ fig : volumes_8_3b ] .
the smallest vison loop in this lattice threads two plaquettes of length 8 and two of length 12 and is visualized in fig .
[ fig : vison_8_3b ] . the calculations for the lattice ( 8,3)b were performed using the following reference gauge : @xmath319 in this gauge , the momentum space hamiltonian reads as @xmath320 where @xmath321 the gauge - fixed matrix representations of the symmetry operators relevant to our classification scheme are @xmath322 and @xmath323 as mentioned in sec
. [ ssec:8b ] , the exchange couplings of the kitaev hamiltonian can be tuned such that the surface fermi arcs touch one another .
as the exchange couplings are tuned away from this point , the fermi arcs split once more , connecting the same pairs of weyl points as before , but winding differently around the surface brillouin zone ( see fig .
[ fig : fermi_arcs_8_3b ] in the main text ) .
this crossing of fermi arcs is pictured in fig . [
fig : surface_crossing ] with @xmath324 and @xmath140 for a slab geometry with 512 layers . also pictured in fig .
[ fig : surface_crossing ] are the probability densities as functions of slab layer of the zero energy states with momentum corresponding to one of the crossing points of the fermi arcs .
the zero modes continue to be highly localized to the surface as the fermi arcs cross , and the splitting / reconnecting of the fermi arcs appears to be a purely surface effect with no corresponding bulk effects .
-surface brillouin zone of lattice ( 8,3)b for a slab geometry 512 layers thick .
_ ( right ) _ probability densities as functions of slab layer of the zero energy states with momentum corresponding to one of the crossing points of the fermi arcs . ]
the lattice ( 8,3)c possesses six loop operators of length 8 and one of length 18 per unit cell . these seven loop operators can be combined to form three closed volumes , each of which must have vanishing total flux , resulting in only four linearly independent loop operators per unit cell .
two of these volumes are constructed only from loops of length 8 and are related to each other by a six - fold screw rotation .
the remaining volume is constructed from six loops of length 8 and two of length 18 .
this larger closed volume and one of the smaller closed volumes are illustrated in fig .
[ fig : volumes_8_3c ] . the smallest vison loop in this
lattice threads four plaquettes of length 8 , visualized in fig .
[ fig : vison_8_3c ] shaded in yellow .
a number of plaquettes of length 18 are also excited two examples of such plaquettes ( shaded in magenta ) are visualized in the second row in fig .
[ fig : vison_8_3c ] .
when considering the assignment of fluxes to the loop operators in this lattice , one might be guided by lieb s theorem and wish to assign @xmath28 flux through the plaquettes of length 8 .
however , such a flux assignment is frustrated due to the volume constraints discussed above . in fig . [
fig : volumes_8_3c](b ) one sees that it takes three such loop operators to form a closed volume .
such a volume must have vanishing total flux , thus making it impossible to assign @xmath28 flux through all plaquettes of length 8 . in the main text
, we thus considered the zero flux sector , the only remaining flux sector that obeys all the lattice symmetries .
the calculations for lattice ( 8,3)c were performed using the following reference gauge @xmath325 in this gauge , the momentum space hamiltonian reads @xmath326 where the matrix @xmath327 is given by @xmath328 and @xmath329 the gauge - fixed matrix representations of the symmetry operators relevant to our classification scheme are @xmath330 and @xmath331 the lattice ( 8,3)n possesses six loop operators of length 8 , four of length 10 , and two of length 12 per unit cell . these twelve loop operators can be combined to form four closed volumes , each of which must have vanishing total flux , resulting in only eight linearly independent loop operators per unit cell .
these closed volumes are illustrated in fig .
[ fig : volumes_8_3n ] .
the smallest vison loop in this lattice threads two plaquettes of length 8 and two of length 10 and is visualized in fig .
[ fig : vison_8_3n ] . the lattice ( 9,3)a has 8 loops per unit cell , all of length 9 . these form two distinct volumes as shown in fig .
[ fig : volumes_9_3a ] where , for the sake of clarity , we only show a small section of the lattice . flipping a @xmath57-type bond
creates a vison loop of length four , visualized in fig .
[ fig : vison_9_3a ] . even though the flux sector considered in the main text is likely not the ground - state sector , note that we can always stabilize this sector as the ground state by adding additional , local terms to the hamiltonian . to that end , we define six 12-loops per unit cell by using the combinations @xmath339 with @xmath184 and @xmath340 with @xmath185 [ see fig . [
fig : volumes_9_3a](a ) for the definition of the loops ] . assigning a negative energy to @xmath28-flux for each of these loops
stabilizes the flux sector considered in the main text .
note that this necessarily implies that the flux through the remaining 12-loop in eq .
thus , assigning @xmath28 flux through all 12-loops is prohibited , i.e. , the z@xmath11 gauge theory in the ( 9,3)a lattice is `` frustrated '' , similarly as for the lattice ( 8,3)c . the lattice ( 10,3)a possesses six loop operators of length 10 per unit cell . these six loop operators can be combined into four closed volumes , each of which must have vanishing total flux , resulting in only two independent loop operators per unit cell .
one of these closed volumes is illustrated in fig .
[ fig : volumes_10_3a ] .
the remaining three volumes are related by a four - fold screw rotation .
the smallest vison loop in this lattice threads ten such plaquettes and is visualized in fig .
[ fig : vison_10_3a ] . the calculations for lattice ( 10,3)a were performed using the following reference gauge : @xmath341 in this gauge , the momentum space hamiltonian reads as @xmath342 where @xmath343 the gauge - fixed matrix representations of the symmetry operators relevant to our classification scheme are @xmath344 the lattice ( 10,3)b possesses four loop operators of length 10 per unit cell .
these four loop operators can be combined to form two closed volumes , each of which must have vanishing total flux , resulting in only two independent loop operators per unit cell .
one of these closed volumes is illustrated in fig .
[ fig : volumes_10_3b ] .
the remaining volume is related by a two - fold screw rotation . the smallest vison loop in this lattice threads six such plaquettes and
is visualized in fig .
[ fig : vison_10_3b ] . the calculations for lattice ( 10,3)b were performed using the following reference gauge @xmath345 in this gauge
, the momentum space hamiltonian reads @xmath346 where @xmath347 the gauge - fixed matrix representations of the symmetry operators relevant to our classification scheme are @xmath348 and @xmath349 the lattice ( 10,3)c possesses three loop operators of length 10 and three of length 12 per elementary 6-site unit cell .
these six loop operators can be combined to form three closed volumes , each of which must have vanishing total flux , resulting in only three linearly independent loop operators per unit cell .
one of these closed volumes is illustrated in fig .
[ fig : volumes_10_3c ] : note that this particular visualization obscures the fact that the loop operators of length 12 are symmetry - related .
the remaining two volumes are related by a six - fold screw rotation .
the smallest vison loop in this lattice threads three plaquettes of length 10 and 11 of length 12 and is visualized in fig .
[ fig : vison_10_3c ] . as mentioned in sec .
[ ssec:10c ] , in order to accommodate the ground - state flux configuration , the unit cell must be enlarged in the 010-direction to a 12 site unit cell .
the sites have been relabeled for the enlarged unit cell , which is depicted in fig .
[ fig:10c_enlarged ] .
the calculations for lattice ( 10,3)c were performed using the following reference gauge : @xmath350 in this gauge , the momentum space hamiltonian reads as @xmath351 where @xmath352 with @xmath353 the gauge - fixed matrix representations of the symmetry operators relevant to our classification scheme are @xmath354 | frustrated quantum magnets can harbor unconventional spin - liquid ground states in which the elementary magnetic moments fractionalize into new emergent degrees of freedom . while the fractionalization of quantum numbers is one of the recurring themes in modern condensed matter physics
, it often remains a challenge to devise a controlled analytical framework tracking this phenomenon .
a notable exception is the exactly solvable kitaev model , in which spin degrees of freedom fractionalize into majorana fermions and a z@xmath0 gauge field .
here we discuss the physics of fractionalization in _ three - dimensional _ kitaev models and demonstrate that the itinerant majorana fermions generically form a ( semi)metal which , depending on the underlying lattice structure , exhibits majorana fermi surfaces , nodal lines , or topologically protected weyl nodes .
we show that the nature of these majorana metals can be deduced from an elementary symmetry analysis of the projective time - reversal and inversion symmetries for a given lattice .
this allows us to comprehensively classify the gapless spin liquids of kitaev models for the most elementary tricoordinated lattices in three dimensions .
we further expand this classification by addressing the effects of time - reversal symmetry breaking and additional interactions . |
we test below the stability of our results to change of model parameter values : results of the calculations by using the same method as in the main text but the other choice of parameters ( @xmath90 and @xmath91 ev ) , which are close to those of ref . .
the results for the temperature dependence of the inverse local magnetic susceptibility are shown in fig .
[ fig : chi_loc2 ] .
we find the crossover discussed in the main text at lower @xmath128 k. the calculation of momentum dependent irreducible susceptibility yields only the uniform ( with respect to @xmath129 ) renormalization without change of qualitative tendencies ( see fig .
[ fig : chi_irr_q2 ] , cf . fig .
[ fig : chi : lda_vs_dmft ] of the main text ) .
we have recalculated exchange interactions from these results and obtain @xmath130 ev vs 0.18 ev in the main text .
this implies lowering of curie temperature , which agrees with approximately renormalization of @xmath30 by 1.5 times ( cf .
[ fig : inverse_chi ] of the main text ) .
the qualitative conclusions of the paper remain unchanged for these parameter values . to obtain exchange interaction , we first determine the bare propagator of magnetic degrees of freedom @xmath132 by requiring that the dressed propagator of @xmath95 field is equal to the susceptibility of itinerant subsystem .
using the random - phase - type approximation , which reduces the orbital- and frequency dependence of the bubble and vertex to the respective single - frequency orbital `` averaged '' quantities , @xmath133 and @xmath86 , we obtain @xmath134where the last term is added to cancel the corresponding bosonic self - energy correction from itinerant degrees of freedom to avoid double - counting , cf .
we represent @xmath135 with momentum - independent @xmath136 ; without loss of generality , we can assume @xmath137 such that @xmath138 from the results of fig . 4 of the main text it follows that @xmath139 . expanding eq .
( [ chis ] ) to first order in @xmath140 , we obtain @xmath141\delta \chi _ { \mathbf{q}}^{\mathrm{0}},\end{aligned}\]]where @xmath142 is the inverse local susceptibility . in practice
, the frequency dependence @xmath143 can be obtained from the dynamic local spin correlation functions , which is characterized by the temperature - independent moment @xmath144 its damping @xmath145 , and the corresponding weiss temperature @xmath146 ( see refs . of the main text ) . since @xmath147ev and
@xmath148ev the momentum dependence is almost cancelled , and we obtain the local bare propagator of spin degrees of freedom,@xmath149considering the renormalization of the propagator @xmath101 by the corresponding boson self
energy corrections ( cf . ref .
@xcite ) , we obtain for the non - uniform susceptibility @xmath150 , \]]which yields eq .
( [ chiq ] ) of the main text ( we use also here that by symmetry @xmath151 ) . | applying the local density approximation ( lda ) and dynamical mean field theory ( dmft ) to paramagnetic @xmath0iron , we revisit a problem of theoretical description of its magnetic properties .
the analysis of local magnetic susceptibility shows that at sufficiently low temperatures @xmath1 k , both , @xmath2 and @xmath3 states equally contribute to the formation of the effective magnetic moment with spin @xmath4 .
the self - energy of @xmath3 states shows sizable deviations from fermi - liquid form , which accompanies earlier found non - quasiparticle form of @xmath2 states . by considering the non - uniform magnetic susceptibility
we find that the non - quasiparticle form of @xmath2 states is crucial for obtaining ferromagnetic instability in @xmath0-iron .
the main contribution to the exchange interaction , renormalized by the effects of electron interaction , comes from the hybridization between @xmath3 and @xmath2 states .
we furthermore suggest the effective spin - fermion model for @xmath0-iron , which allows us to estimate the exchange interaction from paramagnetic phase , which is in agreement with previous calculations in the ordered state within the lda approaches .
elemental iron in its low - temperature body - centered cubic ( bcc ) phase , which is stable below approximately 1200 k , provides unique example of itinerant magnetic @xmath5-electron systems , where formation of well - defined local magnetic moments can be expected .
indeed , the rhodes - wolfarth ratio @xmath6 for this substance is very close to one , which is characteristic feature of systems , containing ( almost ) localized @xmath5electrons ( @xmath7 corresponds to the magnetic moment , extracted from the curie
weiss law for magnetic susceptibility in the paramagnetic phase @xmath8 , and @xmath9 is the saturation moment , @xmath10 is a lande factor , @xmath11 denotes temperature ) . at the same time , the moment @xmath12 has a small fractional part , which is natural for the itinerant material .
this poses natural questions : which electrons mainly contribute to the local - moment spin degrees of freedom of @xmath0-iron ?
what is the appropriate physical model , that describes spin degrees of this substance ? attempting to answer the former question , goodenough suggestedgoodenough , that the @xmath2 electrons are localized , while @xmath3 electrons are itinerant .
this suggestion was later on refined in ref .
, pointing to a possibility , that only some fraction of @xmath2 electrons , contributing to formation of the peak of the density of states near the fermi level , named by the authors as giant van hove singularity , is localized .
( the intimate relation between peaks of density of states and electron localization was also previously pointed out in ref . ) . on contrary , there were statements made that @xmath13 of electrons are localized in iron@xcite .
on the model side , the thermodynamic properties of @xmath0-iron were described within the effective spin @xmath4 heisenberg model @xcite , assuming therefore that the main part of magnetic moment is localized , in agreement with the above - mentioned rhodes - wolfarth arguments .
use of the effective heisenberg model was justified from the ab
initio analysis of spin spiral energies yielding reasonable values of the exchange integrals @xcite .
these considerations however did not take into account strong electronic correlations in @xmath0iron , which important role was emphasized first in ref . .
previous calculations katanin2010,abrikosov2013 within the local density approximation ( lda ) , combined with the dynamical mean - field theory ( dmft ) revealed the presence of non - quasiparticle states formed by @xmath2 electrons , which were considered as a main source of local moment formation in iron , while @xmath3 states were assumed to be itinerant@xcite . at the same time
, magnetic properties of the same @xmath3 states also show some features of local - moment behavior . in particular
, the temperature dependence of inverse local spin susceptibility , which was calculated previously@xcite only at @xmath14 k because of the limitations of hirsch - fye method , is approximately linear , including the contribution of @xmath3 states ; the real part of @xmath3 contribution to dynamic local magnetic susceptibility has a peak at low frequencies , reflecting a possibility of partial local moment formation by @xmath3 states . studying this possibility requires investigation of electronic and magnetic properties at low temperatures , since the energy scale for partially formed local @xmath3 moments can be smaller than for @xmath2 states . although real substance orders ferromagnetically at low temperatures , in the present paper ( as in ref . )
we perform analysis of local properties of iron in the paramagnetic phase to reveal the mechanism of local moment formation . furthermore , we study non - local magnetic susceptibility in the low temperature range @xmath15 @xmath16 which allows us to analyze the mechanism of magnetic exchange . to this end
we use the state of art dynamical mean field theory ( dmft ) calculation with continous time quantum monte - carlo ( ct qmc ) solver@xcite , combined with the ab - initio local density approximation ( lda ) . from our low - temperature analysis
we argue , that @xmath3 electrons are almost equally contribute to the effective local magnetic moment , as the @xmath2 electrons , and play crucial role in the mechanism of magnetic exchange in iron .
in particular , the most important contribution to the exchange integrals comes from the hybridization of @xmath3 and @xmath2 states , which yields _ nearest - neighbour _ magnetic exchange interaction , which agrees well with the experimental data .
we perform the ab initio band structure calculations in lda approximation within tight binding linear muffin tin orbital atomic spheres approximation framework ; the von barth - hedin local exchange - correlation potential @xcite was used .
primitive reciprocal translation vectors were discretized into 12 points along each direction which leads to 72 * k*points in irreducible part of the brillouin zone . for dmft ( ct - qmc ) calculations ,
we use the hamiltonian of hubbard type with the kinetic term containing all @xmath17@xmath18@xmath5 states , being extracted from the lda solution , and the interaction part with density - density contributions for @xmath5 electrons only .
the coulomb interaction parameter value @xmath19 ev and the hund s parameter @xmath20 ev used in our work are the same as in earlier lda+dmft calculations lichtenstein_2001,katanin2010,igoshev2013 . to treat a problem of formation of local moments we consider paramagnetic phase , which is achieved by assuming spin - independent density of states , local self - energy and bath green function . for the purpose of extracting corresponding exchange parameters , we take in lda part physical value of the lattice parameter @xmath21 , corresponding to ferromagnetic state at room temperature .
we consider first the results for the orbital resolved temperature dependent local static spin susceptibility @xmath22 , where @xmath23 is the @xmath24projection of the spin of @xmath5electrons , belonging to the orbitals @xmath25 at a given lattice site @xmath26 , see fig .
[ fig : inverse_chi ] ( for completeness , we also show the total susceptibility @xmath27 which also includes the off - diagonal @xmath3-@xmath2 contribution ) . the temperature dependence of the static inverse local susceptibility is linear ( as was also observed in previous studies lichtenstein_2001,katanin2010,abrikosov2013,igoshev2013 ) , however being resolved with respect to orbital contributions ( see fig .
fig : inverse_chi ) it appears to manifest very different nature of @xmath2 and @xmath3 moments .
the inverse @xmath2 orbital contribution behaves approximately linearly with @xmath11 in a broad temperature rangekatanin2010,abrikosov2013 . at the same time , analyzing low - temperature behaviour , we find that @xmath28 demonstrates a crossover at @xmath29 k between two linear dependences with the low temperature part having higher slope ( i. e. smaller effective moment ) .
note that this feature was not obtained in previous study @xcite because of considering only temperature range @xmath14 k. the scale @xmath30 corresponds to the crossover to non - fermi - liquid behavior of @xmath3 states , see below . and @xmath3 orbital contributions .
dashed lines show linear behavior in different temperature intervals .
, scaledwidth=47.0% ] and @xmath31 in @xmath32iron , extracted from the temperature dependence of local susceptibility , together with the contribution of the @xmath2 and @xmath3 orbitals , scaledwidth=47.0% ] to get further insight into the local magnetic properties of @xmath0-iron , we consider the temperature dependence of the effective magnetic moment @xmath33 and the instantaneous average @xmath34 corresponding to different orbital states , see fig .
[ fig : moms_vs_t ] .
we find , that for @xmath2 electrons both moments saturate at temperatures @xmath1 k and remain approximately constant up to sufficiently low temperatures . comparing the value of the square of the moment @xmath35 , extracted from the curie - weiss law for local susceptibility , and the instantaneous average @xmath36 with the corresponding filling @xmath37 , we find that the major part of @xmath2 electrons determine the instantaneous average @xmath38 , and at least half of them contribute to the sufficiently long - living ( on the scale of @xmath39 ) local moments . at the same time , for @xmath3 electronic states the abovementioned crossover between the high - temperature value @xmath40 and the low temperature value @xmath41 is present , which , comparing to @xmath42 shows that at least @xmath43 of @xmath3 electrons participate in the effective local moment formation at low temperatures . yet , the corresponding low - temperature effective moments @xmath44 and @xmath45 are comparable ( each of them is approximately @xmath46 corresponding to the effective spin @xmath47 ) , showing important role of @xmath3 electrons in the formation of the total spin @xmath4 state . .
black solid lines illustrate the results of calculations at temperatures ( from top to bottom ) @xmath48 ev .
the green dot dashed line present fits to the the fermi - liquid dependence in the range @xmath49 ev , while blue dashed lines present fits to the non - fermi - liquid dependence ( see text ) .
dots denote matsubara frequencies @xmath50.,scaledwidth=44.0% ] although the self - energy calculations @xcite yield quasiparticle - like form of @xmath3 electron self - energy , the low - frequency and low - temperature dependence of self - energy shows pronounced deviations from the fermi - liquid behavior , see fig . [ fig : sigma ] . to analyse the frequency dependence of the self - energy on imaginary frequency axis , we fit the obtained results by the fermi - liquid dependence @xmath51\nu + \sigma ( t)\nu ^{2},$ ] where @xmath52 is the damping of electrons at the fermi level , @xmath53 is the temperature - dependent quasiparticle residue . alternatively , we consider the fit @xmath54 with some exponent @xmath55 .
the latter dependence corresponds to the non - fermi - liquid behavior of @xmath3 electrons .
the obtained results are presented in the table .
@xmath56 the linear - quadratic fits are applicable only at @xmath57 ev ; at sufficiently small @xmath58 they also do not fit the obtained results well .
we find that the spectral weight @xmath53 pronouncely decreases with decrease of temperature , and the coeffitient @xmath52 obviously does not obey the fermi - liquid dependence @xmath59 .
these observations show that sizable deviations from fermi - liquid picture can be expected .
the power - law fits yield much better agreement in a broad range of frequencies @xmath60 ev , describing at the same time correctly the low - frequency behavior .
the coefficients @xmath61 @xmath62 of these fits show very weak temperature dependence ( the contribution @xmath63 is almost negligible ) , while the damping @xmath64 and the exponent @xmath0 slightly decrease with temperature , being related by @xmath65 .
these observations imply that @xmath3 electronic subsystem is better described by non - fermi liquid behavior at low temperatures , which reflects its participation in the formation of local moments in @xmath0-iron .
remarkably , consideration of the three - band model in ref
. showed similar dependence of the self - energy @xmath66 due to hund exchange interaction , which allows to attribute the @xmath3 subsystem in iron as close to the " spin freesing transition , accroding to the terminology of ref . .
to get further insight into the formation of effective local moments and extract corresponding exchange integrals , we calculate the momentum dependence of particle - hole bubble @xmath67 , which is obtained using paramagnetic lda and lda+dmft electronic spectrum [ @xmath68 is the corresponding electronic green function for the transition from the orbital state @xmath69 to @xmath70 , @xmath71 is a fermionic matsubara frequency ; for more details on the calculation procedure see ref . ] . at @xmath72 k calculated in high symmetry directions of the brillouin zone .
the contributions @xmath73 , @xmath74 , @xmath75 , and the hybridization part @xmath76 are shown by black , red , green and blue lines , respectively .
solid ( dashed ) lines correspond to lda+dmft ( lda ) results .
the lda+dmft estimate for @xmath77 is shown by magenta short dashed line .
, scaledwidth=44.0% ] the results for lda and lda+dmft approaches at @xmath72 k are presented in the fig .
[ fig : chi : lda_vs_dmft ] ( we find that the lda+dmft results are almost temperature - independent at low @xmath11 ) . for the bubble , calculated using purely lda spectrum ( i.e. with the assumption that all electrons are itinerant ) , the maximum of @xmath78 is located at the point @xmath79 , while the ferromagnetic instability in @xmath0-iron requires maximum of @xmath78 at @xmath80 and low @xmath81 if one neglects the non - local vertex corrections .
one can observe , that the main contribution to this incorrect behavior of the bubble originates from the @xmath2 electron part , @xmath82 .
both @xmath82 and @xmath83 contributions are however strongly influenced by the account of the local self energy corrections to the green s function in dmft approach , which correspond physically to account of partial localization of @xmath5-electrons .
these corrections mainly change @xmath84 and shift the maximum of @xmath85 to @xmath86 point ( @xmath80 ) .
note that within lda+dmft , intra
orbital contributions to @xmath82 and @xmath87 are only weakly momentum - dependent ; they also behave similarly , varying counter phase . according to the general ideas of spin - fluctuation theory @xcite
, this weak momentum dependence can be ascribed to the formation of the effective moments from @xmath2 and @xmath3 states . in agreement with the abovediscussed consideration
, the @xmath84 contribution has even weaker dispersion than the @xmath88 part . at the same time ,
strongly dispersive @xmath89 contribution , which is assumed to correspond to the ( remaining ) itinerant degrees of freedom , provides the maximum of the resulting @xmath78 at @xmath80 and appears to be the main source of the stability of the ferromagnetic ordering in iron within lda+dmft approximation .
the obtained results do not change qualitatively for the other choice hubbard interactions ( as we have verified for @xmath90 and @xmath91 ev ) , see supplementary material @xcite . to see the quantitative implications of the described physical picture
, we consider the effective spin - fermion model @xmath92 c_{l^{\prime } \sigma } ( \mathrm{i}\nu _ { n } ) \notag\end{aligned}\]](@xmath93 is a bosonic matsubara frequency , @xmath94 combines site and orbital indices ) , describing interaction of itinerant electrons with ( almost ) _ local _ spin fluctuations ( in contrast to critical spin fluctuation in cuprates @xcite ) , see also ref .
@xcite .
we assume here that the coulomb and hund s interaction acting within @xmath2 and @xmath3 orbitals results in a formation of some common local moment ( field @xmath95 ) , while the remaining itinerant degrees of freedom are described by the field @xmath96 formed from the grassmann variables @xmath97 @xmath98 and @xmath99 are the hamiltonian and local self - energy corrections to the lda spectrum ( the latter is assumed to be local and therefore diagonal with respect to orbital indices ) .
the interaction between the two subsystems ( localized and itinerant ) , which are formed from the @xmath5electronic states , is driven by hund s constant coupling @xmath100 . considering the renormalization of the propagator @xmath101 by the corresponding boson self
energy corrections , we obtain for the non - uniform susceptibility ( see supplementary material @xcite ) @xmath102where @xmath103 is the local spin susceptibility and @xmath104 is the exchange interaction , which fulfills @xmath105 ( no spin self - interaction ) .
we find @xmath106 , @xmath107 $ ] is the intra - orbital part , while @xmath108 results from the hybridization of states of different symmetry .
the contribution @xmath109 is approximately twice smaller than @xmath110 and therefore the main contribution to the magnetic exchange comes from the hybridization of @xmath3 and @xmath2 states .
the whole momentum dependence of @xmath111 can be well captured by the nearest
neighbor approximation for effective exchange integrals only , @xmath112 while @xmath109 has more complicated momentum dependence . restricting ourselves by considering the contribution @xmath113 , ( we assume that the contribution @xmath109 is further suppressed by the non - local and vertex corrections ) , from fig . [
fig : inverse_chi ] we find at @xmath72 k the value @xmath114 ev .
this value , as well as the momentum dependence of @xmath111 agrees well with the result of s.v .
okatov et al .
@xcite . the obtained results together with @xmath115 ( see fig .
2 ) provide an estimate for the curie temperature ( we assume @xmath116 ) , which can be obtained from the divergence of @xmath117 : @xmath118and appears comparable with the result of full dmft calculation , and therefore shows that the above model is adequate for describing magnetic properties of the full @xmath119-band hubbard model .
( note that the overestimation of @xmath120 in dmft approach in comparison with the experimental data is due to density - density approximation for the coulomb interactionanisimov and ( to minor extent ) due to presence of non - local fluctuations , not accounted by dmft ) . neglecting longitudinal fluctuations of field @xmath95
we can map the model ( [ sf ] ) to an effective @xmath4 heisenberg model @xmath121 to estimate the spin wave spectrum : @xmath122we obtain the corresponding spin stiffness @xmath123@xmath124 in a good agreement with the experimental data @xmath125@xmath124 ( ref . ) . in conclusion
, we have considered the problem of the description of effective local moments in @xmath0-iron based on the electronic spectrum in paramagnetic phase within lda+dmft approximation .
we find that local moments are formed by both @xmath2 and @xmath3 orbital states , each of them contributing a half of the total moment @xmath126 for @xmath3 electronic states we find pronounced features of non - fermi - liquid behavior , which accompanies earlier observed non - quasiparticle form of @xmath2 states .
the local moment and itinerant states interact with itinerant states via hund interaction , yielding magnetic exchange between the local - moment states via the effective rkky type mechanism .
the obtained exchange integrals are well captured by the lda+dmft approach .
the main origin of the intersite interaction of these moments is attributed to the @xmath2-@xmath3 hybridization , which yields magnetic exchange , dominating on the nearest - neighbour sites .
contrary to the previous studieslichtenstein1987,gornostyrev , we do not however assume some magnetic ordering for the electronic system .
we also emphasize that non
local self energy corrections , as well as vertex corrections , missed in our investigation , can make the described physical picture more precise .
in particular , non - local effects allow for the non zero non - diagonal @xmath2@xmath3 self - energy matrix elements and therefore possibly renormalize the strength of exchange interaction , as well as the self - energy of @xmath3 electronic states .
the role of the vertex corrections , only roughly accounted in the considered approach , also requires additional study .
therefore further investigation using powerful theoretical techniques of dynamic vertex approximation @xcite , dual fermion@xcite , or other non - local approaches is of certain interest .
the authors are grateful to yu .
n. gornostyrev , a. v. korolev , and k. held for useful discussions .
the work of p. a. igoshev was supported by the russian foundation for basic research ( project no .
14 - 02 - 31603 ) and act 211 government of the russian federation 02.a03.21.0006 ; a. a. katanin acknowledges support of the program of `` dynasty '' foundation .
the calculations were performed using " uran supercomputer of imm ub ras .
99 j. b. goodenough , phys . rev . * 120 * , 67 ( 1960 ) .
v. yu . irkhin , m. i. katsnelson , and a. v. trefilov , j. phys . : condens .
matter * 5 * , 8763 ( 1993 ) .
s. v. vonsovskii , m. i. katsnelson , and a. v. trefilov , fiz . met .
metalloved .
76 ( 3 ) 3 ( 1993 ) ; 76 ( 4 ) , 3 ( 1993 ) .
m. b. stearns , phys .
rev .
b * 8 * , 4383 ( 1973 ) ; r. mota and m. d. coutinho - filho , phys . rev .
b * 33 * , 7724 ( 1986 ) .
f. krmann , a. dick , b. grabowski , b. hallstedt , t. hickel , and j. neugebauer , phys .
rev .
b * 78 * , 033102 ( 2008 ) s.v .
okatov , yu.n .
gornostyrev , a.i .
lichtenstein , and m.i .
katsnelson , phys .
rev .
b * 84 * , 214422 ( 2011 ) .
a. i. lichtenstein , m. i. katsnelson , and g. kotliar , phys .
rev .
lett . * 87 * , 067205 ( 2001 ) .
a. a. katanin , a. i. poteryaev , a. v. efremov , a. o. shorikov , s. l. skornyakov , m. a. korotin , v. i. anisimov , phys .
rev .
b * 81 * , 045117 ( 2010 ) .
p. werner et al . , phys .
rev .
lett . * 97 * , 076405 ( 2006 ) .
l. v. pourovskii , t. miyake , s. i. simak , a. v. ruban , l. dubrovinsky , and i. a. abrikosov , phys .
rev .
b * 87 * , 115130 ( 2013 ) u. von barth and l. hedin , j. phys .
c 5 , 1629 ( 1972 ) .
p. werner , e. gull , m. troyer , and a.j .
millis , phys .
rev .
lett .
* 101 * , 166405 ( 2008 ) .
p.a .
igoshev , a.v .
efremov , a.i .
poteryaev , a.a .
katanin , v.i .
anisimov , phys .
rev .
b * 88 * , 155120 ( 2013 ) .
t. moriya , spin fluctuations in itinerant magnets .
springer - verlag , berlin , heidelberg , 1985 .
see supplementary material at http:// jrg schmalian , david pines , and branko stojkovi , phys .
rev .
b * 60 * , 667 ( 1999 ) ; a. abanov , a. v. chubukov , and j. schmalian , adv . phys . * 52 * , 119 ( 2003 ) .
a. a. katanin , a. toschi , and k. held , phys .
rev .
b * 80 * , 075104 ( 2009 ) . v. i. anisimov , a. s. belozerov , a. i. poteryaev , and i. leonov , phys . rev .
b * 86 * , 035152 ( 2012 ) .
a. i. liechtenstein , m. i. katsnelson , v. p. antropov , v. a. gubanov , jmmm * 67 * , 65 ( 1987 ) .
h.a .
mook and r.m .
nicklow , phys .
rev .
b * 7 * , 336 ( 1973 ) .
see , e.g. a. toschi , a. a. katanin , and k. held , phys .
rev .
b * 75 * , 045118 ( 2007 ) ; a. toschi , g. rohringer , a. a. katanin , k. held , ann .
der phys . , 523 , * 698 * ( 2011 ) .
see , e.g. , a. n. rubtsov , m. i. katsnelson , a. i. lichtenstein , a. georges , phys.rev .
b * 79 * 045133 ( 2009 ) . |
the exponential expansion of space in inflationary cosmology leads to the stretching of modes which were in the trans - planckian regime at the beginning of inflation into the observable range .
this leads to the possibility , first raised explicitly in @xcite , that trans - planckian physics might be observable today in the cosmic microwave background . in earlier work
@xcite we addressed this issue in a simple toy model obtained by replacing the linear dispersion relation of the cosmological fluctuations by new dispersion relations which differ from the linear one on length scales smaller than the planck length ( the same dispersion relations had been used earlier @xcite in the context of an analysis of possible trans - planckian effects on black hole radiation ) .
we were able to construct dispersion relations which give rise to large ( order one ) corrections to the usual spectrum of fluctuations , but the price to pay is a fine - tuning of the parameters describing the model and/or a back - reaction problem
. this question has been further analyzed in many papers ( see for instance refs .
it was found that in order to obtain measurable differences in the predictions , non - adiabatic evolution of the state on trans - planckian scales is required . in another line of approach to the _ trans - planckian challenge _ to inflationary cosmology ,
the possibility of measurable effects of trans - planckian physics on observables such as cmb anisotropies and power spectra of scalar and tensor metric fluctuations was studied @xcite in models where the trans - planckian physics is based on stringy space - time uncertainty relations .
in particular , the authors of @xcite found a spectrum with oscillations of amplitude @xmath4 , where @xmath5 is the hubble parameter during inflation and @xmath1 a characteristic scale at which the trans - planckian physics shows up , superimposed on the usual scale - invariant spectrum , whereas the authors of @xcite found only much smaller effects .
the trans - planckian problem was also tackled in the framework of non - commutative geometry in ref .
it was found that the effect is of order @xmath6 .
it was also shown in this article that non - commutative geometry implies the presence of a preferred direction which would result in a correlation between different multipoles @xmath7 and @xmath8 . in yet another approach to the trans - planckian issue , danielsson @xcite ( see also ref .
@xcite ) suggested to replace the unknown physics on trans - planckian scales by assuming that the modes representing cosmological fluctuations are generated mode by mode at the time when the wavelength of the mode equals the planck length , or more generally when it equals the length @xmath0 associated with the energy scale @xmath1 of the new physics which sets the initial conditions .
there is a one - parameter family of vacuum states ( @xmath9 vacua ) of a free quantum field in de sitter space which can be considered , and for nontrivial @xmath9 vacua danielsson found effects of the choice of the initial state which are of linear order in the ratio @xmath2 also occur in models of power - law inflation . ] , and such effects could be seen in observations @xcite .
vacua are problematic from the point of view of interacting quantum field theory @xcite .
however , very recently it has been shown @xcite how to define interacting quantum field theory about an @xmath9 vacuum . ]
similar results were found by easther et al .
@xcite , whereas niemeyer et al .
@xcite have argued that if the modes are set off in the initial adiabatic vacuum state when their wavelength equals @xmath0 , then the effects are of order @xmath10 and hence ( in usual models ) completely negligible . using an effective field theory method , kaloper et al .
@xcite have argued that the effects of trans - planckian physics on cosmological fluctuations should be at most of the order @xmath11 , assuming that the modes are in the adiabatic vacuum state when the wavelength is equal to the hubble radius ( see ref .
@xcite for a criticism of imposing initial conditions at hubble radius crossing , and see ref .
@xcite for counterexamples to the claims of ref .
@xcite ) . in this paper , we re - consider the calculation of the spectrum of cosmological perturbation in the _ minimal trans - planckian _ setting @xcite when mode by mode the initial conditions for the mode are set when the wavelength equals the planck length ( or , more generally , the length scale of the new physics ) .
we find that the overall amplitude of the correction terms ( compared to the usual spectra ) depends sensitively on the precise prescription of the initial state , it depends on whether one is studying power - law or slow - roll inflation , and it also depends on whether one is computing the spectrum of scalar metric fluctuations or of gravitational waves .
some of the `` discrepancies '' between the results of previous studies is due to the fact that different quantities were calculated in different models .
we show that when the initial state is chosen to be the instantaneous minkowski vacuum , then the deviations of the power spectrum from the usual result are of the order @xmath10 , in agreement with what was found in @xcite . in an arbitrary @xmath12 vacuum
, the choice of the value of @xmath9 has an effect on the amplitude of the fluctuation spectrum which is not suppressed by any power of @xmath2 .
however , if @xmath9 is independent of @xmath13 , the effect will not be distinguishable from a slight change in the underlying parameters of the background inflationary model .
however , in general ( and specifically in the choice of the vacuum made in @xcite , the amplitude of the correction term in the power spectrum will have a k - dependent ( and hence observable ) piece which is first order in @xmath2 , at least in the case of the spectrum of gravitational waves .
while this paper was being finalized , three preprints appeared which investigate related aspects of the trans - planckian problem . in ref .
@xcite , the choice of various initial states was related to the minimization of different hamiltonians . in ref .
@xcite , the predictions of inflation for the power spectrum of fluctuations was studied for a two parameter class of initial states , and the amplitude of the corrections compared to the usual results was seen to depend sensitively on which state is chosen . in ref .
@xcite , the fact that the definitions of the bunch - davies and of the adiabatic vacua have some intrinsic ambiguities in a universe with a de sitter phase of finite duration was analyzed .
the line element for the spatially flat friedmann - lemaitre - robertson - walker ( flrw ) background plus small perturbations can be written as @xcite : @xmath14{\rm d}x^i{\rm d}x^j\}\ , , \end{aligned}\ ] ] where the functions @xmath15 , @xmath16 , @xmath17 and @xmath18 represent the scalar sector whereas the traceless and transverse tensor @xmath19 ( @xmath20 ) , represents the tensor sector , i.e. the gravitational waves .
the time @xmath21 is the conformal time and is related to the cosmic time @xmath22 by the relation @xmath23 .
it is convenient to introduce the background quantity @xmath24 defined by @xmath25 , where a dot means differentiation with respect to cosmic time and @xmath26 is the hubble rate , @xmath27 . using the conformal time we may rewrite @xmath28 as @xmath29 , where @xmath30 , and a prime denotes differentiation with respect to the conformal time . for power - law scale factors , @xmath31 , the function @xmath24 turns out to be a constant , @xmath32 , which vanishes in the particular case of the de sitter space - time characterized by @xmath33 .
the perturbed einstein equations provide us with the equations of motions for the cosmological perturbations . at the linear level , each type of perturbations decouple and we can treat them separately . in the tensor sector
( which is automatically gauge - invariant ) we define the quantity @xmath34 for each mode @xmath13 according to @xmath35 where @xmath36 is the polarization tensor .
the plane waves appear in the previous expression because the space - like sections are flat . at linear order , gravitational waves
do not couple to matter and , as a consequence , the equation of motion is just given by the perturbed vacuum einstein equation .
explicitly , it reads @xcite : @xmath37 this equation is similar to a time - independent schrdinger equation with an effective potential given by @xmath38 .
it can also be viewed as the equation of a parametric oscillator whose time - dependent frequency is given by @xmath39 , see ref .
@xcite . in the vacuum state ,
the two point correlation function of gravitational waves reads @xmath40 the power spectrum @xmath41 which appears in the expression for the two point correlation function is given by @xmath42 this quantity is _ a priori _ time and wavenumber dependent but for super - horizon modes , it turns out to be time - independent because the growing mode is given by @xmath43 .
let us now turn to scalar metric ( density ) perturbations .
the two most important differences with the gravitational waves are that the scalar sector is gauge - dependent and that the scalar perturbations of the metric are coupled to the perturbations of the stress - energy tensor describing the matter .
scalar perturbations of the geometry can be characterized by the two gauge - invariant bardeen potentials @xmath44'$ ] and @xmath45 @xcite . from now on ,
we restrict ourselves to the case where the matter is described by a scalar field : @xmath46 .
fluctuations in the scalar field are characterized by the gauge - invariant quantity @xmath47 .
the full set of the scalar perturbed ( gauge - invariant ) einstein equations is @xmath48 , \quad \partial _
i({\cal h}\phi + \psi ' ) = \frac{\kappa } { 2}\varphi _ 0'\partial _
i\delta \varphi \ , , \\ & & \partial _ i\partial ^j(\phi -\psi ) = 0\ , , \quad ( 2{\cal h}'+{\cal h}^2)\phi + { \cal h}\phi ' + \psi ' ' + 2{\cal h}\psi ' -\frac{1}{3}\partial _ k\partial ^k(\phi -\psi ) = \frac{\kappa } { 2 } \biggl[-(\varphi _
0')^2\phi + \varphi _ 0'\delta \varphi ' -a^2\frac{{\rm d}v}{{\rm d}\varphi _ 0}\delta \varphi \biggr]\ , , \end{aligned}\ ] ] where @xmath49 , @xmath50 being the planck mass . at this point ,
one has to be careful because the case @xmath51 , which corresponds to the de sitter space - time , is particular .
in this situation , the solution of the above system of equations is simply @xmath52 , i.e. there are no density perturbations at all ( it is also necessary to require that the bardeen potential is finite at infinity ) .
if @xmath53 , then everything can be reduced to the study of a single gauge - invariant variable ( the so - called mukhanov - sasaki variable ) defined by @xcite @xmath54 it turns out to be more convenient to work with the variable @xmath55 defined by @xmath56 .
its equation of motion is very similar to that of the gravitational waves and reads @xcite : @xmath57\mu _ { _ { \rm s}}=0\ .\ ] ] some remarks are in order here .
firstly , let us stress again that the above equation can be used only if @xmath53 or @xmath58 because in course of its derivation , we have divided by @xmath59 . secondly ,
the effective potential for density perturbations , @xmath60 is different from the potential for gravitational waves .
equivalently , its time - dependent frequency is also different and given by @xmath61 .
thirdly , in the case of power - law inflation , the function @xmath28 is a constant and we have @xmath62 .
fourthly , density perturbations are often characterized by the so - called conserved quantity @xmath63 defined by @xmath64 , where @xmath65 is the equation of state parameter , i.e. the ratio of pressure to energy density of the background scalar field .
this quantity is directly linked to @xmath55 through the equation @xmath66 .
the integration of eq .
( [ eomscalar ] ) leads to the primordial spectrum of this quantity , namely @xmath67 the fact that this quantity is meaningless for the de sitter case , @xmath68 , is obvious .
let us now consider a different situation : there are no cosmological fluctuations anymore but just a test scalar field @xmath69 living in a flrw universe .
if we fourier expand the scalar field according to @xmath70 then the klein - gordon equation takes the form @xmath71 we recognize the equation of motion of gravitational waves .
the two - point correlation function of the scalar field can be written as @xmath72 the quantity @xmath73 which appears in the expression for the two point correlation function is the power spectrum of the scalar field and is given by @xmath74 we also see that the equation of motion for density perturbations is equivalent to the equation for a scalar field if the function @xmath75 is a non - vanishing constant .
this is why , for this class of models , the study of density perturbations is in fact equivalent to the study of a scalar field .
however , there is an important exception to this prescription : the de sitter space - time since , in this case , @xmath76 so that the equation of motion is not given by eq .
( [ eomscalar ] ) as already mentioned above .
therefore , it is inconsistent to first assume that @xmath28 is a constant , then to use eq .
( [ specdp ] ) in order to calculate the power spectrum and finally to particularize the result to the de sitter case .
this procedure can only lead to the determination of the spectrum of gravitational waves and/or of a test scalar field but not of density perturbations in the de sitter space - time .
the case of power - law inflation , for which the scale factor can be written as @xmath77 , is important because the equation of motion for the cosmological perturbations can be solved explicitly .
the parameter @xmath78 is a length since we have chosen to work with a dimension - full scale factor . since the function @xmath79 reduces to a constant , the effective potential for density perturbations simplifies to @xmath80 and becomes identical to the gravitational waves effective potential
. its explicit form reads @xmath81 .
then , the exact solution for the variables @xmath82 , @xmath34 and @xmath83 reads @xmath84\ , .\ ] ] in the above expression , @xmath85 is a bessel function of order @xmath86 . the two @xmath13-dependent constants @xmath87 and @xmath88 are fixed by the initial conditions .
we now consider the standard calculation of inflationary cosmology . in this case , the initial conditions are fixed in the infinite past .
when @xmath89 , the mode function tends to a plane wave with positive and negative frequency .
the usual procedure requires that @xmath90 let us notice that the origin of the minus sign for density perturbations is the minus sign in the relation @xmath91 , @xmath92 being the quantity which is canonically quantized , i.e. which behaves as @xmath93 in the ultraviolet limit .
then , the constants @xmath87 and @xmath88 are completely specified and are given by @xmath94 this implies that the mode functions @xmath95 can be expressed as @xmath96 where @xmath97 is the hankel function of first kind ] one can easily check that the mode function @xmath98 has indeed the required limit given by eq .
( [ stic ] ) . ] .
this result allows us to calculate the power spectra . for density perturbations and gravitational waves ,
one respectively finds , on super - hubble scales @xmath99 where @xmath100 is the planck length and the function @xmath101 is given by @xmath102 ^ 2\ , .\ ] ] in the above definition of the function @xmath101 , @xmath103 denotes euler s integral of the second kind . for the de sitter case , one has @xmath104 .
however , for this case the amplitude of density perturbations blows up since @xmath105 while the amplitude of the gravitational waves spectrum remains finite .
the origin of the singular limit for density perturbations is again the factor @xmath106 at the denominator of the expression for the scalar power spectrum .
in fact , this case must be analyzed separately and one can show that density perturbations do not exist in de sitter space - time .
we see that power - law inflation leads to @xmath107 with @xmath108 .
explicitly , one has @xmath109 the case @xmath33 leads to a scale - invariant spectral index , namely @xmath110 and @xmath111 .
the power - spectrum of the scalar field can be also calculated and the result reads ( contrary to the previous power spectra , the power spectrum of the scalar field is a dimension - full quantity ) @xmath112 it is also convenient to expressed this power spectrum in terms of the hubble parameter during inflation @xmath113 $ ] .
this permits to replace the scale @xmath78 in the above equation and leads to @xmath114 of course , in spite of the fact that the time dependence now appears explicitly , the power spectrum remains a time independent quantity . in the de sitter case ,
@xmath33 , one recovers the result often cited in the literature , namely @xmath115 ^ 2 $ ] .
the spectrum becomes scale - invariant and its amplitude remains finite .
the situation , except for some unimportant numerical factors , is very similar to that of gravitational waves . to conclude this subsection ,
let us emphasize that the previous discussion has shown that the analogy between density perturbations and a scalar field must be used cautiously .
the spectral indices are similar even in the de sitter limit but the amplitudes differ radically in this limit .
the slow - roll method is an approximation scheme which allows us to go beyond the simple power - law solutions considered in the previous section .
it permits to treat a more general class of inflaton potentials . at leading order ,
the approximation is controlled by the slow - roll parameters ( see e.g. ref .
@xcite ; for a new set of slow - roll parameters with a very nice interpretation , see ref .
@xcite ) defined by : @xmath116 the quantity @xmath117 is the inflaton potential .
we see that @xmath118 , where @xmath28 is the function that has been introduced before .
the slow - roll conditions are satisfied if @xmath119 and @xmath120 are much smaller than one and if @xmath121 .
since the equations of motion for @xmath119 and @xmath120 can be written as : @xmath122 it is clear that this amounts to considering @xmath119 and @xmath120 as constants if one works at first order in the slow - roll parameters .
this property turns out to be crucial for the calculation of the power spectra of cosmological perturbations . for power - law inflation
the slow - roll parameters satisfy : @xmath123 , @xmath124 .
the slow - roll approximation can be viewed as a kind of expansion around the de sitter space - time . indeed , at leading order , one has @xmath125 which implies that the scale factor behaves like @xmath126 .
interestingly enough , the effective power index at leading order depends on @xmath119 only . at this point
, one should make the following remark .
using the relation @xmath32 , the scale factor of power - law inflation can be re - written as @xmath127 . from this expression
, one might be tempted to write that , in the slow - roll framework , @xmath128 since @xmath129 in this approximation .
of course , this expression is inconsistent because it contains an infinite number of power of @xmath130 .
what should be done is to expand the expression @xmath131 in terms of @xmath119 from which we recover that the scale factor is given by @xmath126 . as long as one decides to keep high order terms , the whole hierarchy of slow - roll parameters should enter the game
this shows that the slow - roll approximation does not only consist in naively expanding the power index of the scale factor in powers of @xmath130 .
this conclusion is reinforced by a study of the cosmological perturbations within this approximation .
the effective potential of density perturbations can be calculated exactly in terms of the slow - roll parameters .
the result is : @xmath132 \ .\ ] ] we have seen before that , in the slow - roll approximation , @xmath133 .
this implies that the effective potential reduces to @xmath134 since , at leading order , @xmath119 and @xmath120 must be seen as constants in the slow - roll approximation , the equation of motion is of the same type as in power - law inflation and the solution is expressed in terms of bessel functions according to : @xmath135.\ ] ] the parameter @xmath136 appearing in the order of the bessel function is given by @xmath137 a comment is in order here : the potential @xmath138 depends on the scale factor and its derivatives only and we have seen before that the scale factor behaves as @xmath139 .
therefore , one might think that @xmath138 should depend on @xmath119 only .
this is not the case .
the reason is that @xmath140 contains terms like @xmath141 ( for instance ) which are linear in @xmath120 .
first one must calculate all derivatives , replace them with their expression in terms of @xmath119 and @xmath120 , and only then consider that the slow - roll parameters are constant .
for gravitational waves , the same lines of reasoning can be applied .
the effective potential can be written as @xmath142 and gives in the slow - roll limit @xmath143 therefore , the solution of @xmath34 is similar to the one given in eq .
( [ regionii ] ) , where the effective index of the bessel function is now given by : @xmath144 .
this time , the spectral index only depends on @xmath130 as expected from the shape of the tensor effective potential .
fixing the initial conditions in the infinite past in the standard manner , we arrive at the following expressions for the power spectra @xmath145\ , , \\
\label{srgw } { \cal p}_h(k ) & = & \frac{16 h^2}{\pi m_{_{\rm pl}}^2 } \left[1 - 2(c+1)\epsilon - 2\epsilon \ln \left(\frac{k}{k_*}\right)\right]\ , , \end{aligned}\ ] ] where @xmath146 , @xmath147 being the euler constant .
all quantities are evaluated at hubble radius crossing .
the scale @xmath148 is called the pivot scale , see refs .
@xcite and @xcite .
one sees again that the de sitter limit @xmath149 is ill - defined for scalar metric fluctuations , whereas it is well - defined for gravitational waves .
we now consider the cosmological perturbations of quantum - mechanical origin in the framework of the minimal trans - planckian physics .
the new ingredient consists in assuming that the fourier modes never penetrate the trans - planckian region .
rather , the main idea is that a fourier mode is `` created '' when its wavelength becomes equal to a new fundamental characteristic scale @xmath150 .
then , the evolution proceeds as usual since the equations of motion of the mode functions @xmath95 are taken to be unmodified and still given by eqs .
( [ eomtensor ] ) and ( [ eomscalar ] ) . in this way
, the changes are entirely encoded in the initial conditions and there is no need to postulate some ad - hoc trans - planckian physics .
the time of `` creation '' of the mode of comoving wavenumber @xmath13 , @xmath151 , can be computed from the condition @xmath152 which implies that @xmath151 is a function of @xmath13 .
this has to be compared with the standard calculations where the initial time is taken to be @xmath153 for any fourier mode @xmath13 ( in a certain sense , this initial time does not depend on @xmath13 ) .
the situation is summarized in fig .
[ tpl3 ] in the framework described above , a crucial question is in which state the fourier mode is created at the time @xmath151 .
there is now no asymptotic region in the infinite past where the standard prescriptions can be applied . in this article
, we consider the most general conditions , namely that the mode is placed in an @xmath154-vacuum according to @xmath155 where @xmath156 is the effective frequency for density perturbations and gravitational waves .
the coefficients @xmath157 and @xmath158 are _ a priori _ two arbitrary complex numbers satisfying the condition @xmath159 .
the instantaneous minkowski state corresponds to @xmath160 and @xmath161 . if , in addition , we take @xmath153 , we recover the standard choice , see eq .
( [ stic ] ) since @xmath162 .
we are now in a position where we can compute the mode functions and the corresponding power spectra for density perturbations and gravitational waves . in this section ,
we write the scale factor as @xmath163 , where @xmath164 is a generalized index defined by @xmath165 since the initial conditions given by eqs .
( [ ci1 ] ) and ( [ ci2 ] ) are different from the standard ones , we will obtain different mode functions .
the new mode functions that we want to calculate can be expanded according to @xmath166^*\ , , \ ] ] where @xmath167 denotes the mode functions obtained in the standard situation and given by the expression @xmath168 in this equation , we have introduced another generalized index @xmath169 defined by @xmath170 for power - law inflation the generalized index is the same for density perturbations and for gravitational waves ( hence we do not need to distinguish them in the above definition ) . for slow - roll inflation , as already mentioned above , they differ .
the coefficients @xmath171 and @xmath172 are readily obtained using the initial conditions given in eqs .
( [ ci1 ] ) and ( [ ci2 ] ) @xmath173 + \biggl[1 + 2i\frac{c_k - d_k}{c_k+d_k}\omega _ { _ { \rm s , t}}(\eta _ k)\eta _ k \biggr]h^{(2)}_{1/2-\nu}\biggr\}\ , , \\
\label{beta } \beta _ { _ { \rm s , t } } ( k ) & = & \frac{1}{4}(c_k+d_k ) { \rm e}^{-i\pi \nu/2 } \sqrt{\frac{\pi } { -2\omega _ { _ { \rm s , t}}(\eta _ k)\eta _ k } } \biggl\{k\eta _ k\biggl[h^{(1)}_{3/2-\nu}-h^{(1)}_{-1/2-\nu}\biggr ] + \biggl[1 + 2i\frac{c_k - d_k}{c_k+d_k}\omega _ { _ { \rm s , t}}(\eta _ k)\eta _ k \biggr]h^{(1)}_{1/2-\nu}\biggr\}\ , , \end{aligned}\ ] ] where the hankel functions are evaluated at @xmath174 .
the knowledge of the coefficients @xmath171 and @xmath172 is equivalent to the knowledge of the modified mode function , see eq .
( [ nonstmodef ] ) .
the quantity @xmath174 can be written as @xmath175 where @xmath176 .
the explicit expression of @xmath177 can be easily derived .
writing that @xmath178 , we find that @xmath179 where the time @xmath180 is a given , _ a priori _ arbitrary , time during inflation which , and this is the important point , does not depend on @xmath13 .
the quantity @xmath181 is defined by @xmath182 . using that @xmath183 and @xmath184 we finally arrive at @xmath185 in the case of slow - roll inflation , the previous expression reduces to @xmath186^{-\epsilon } $ ] . for the de sitter case ,
@xmath187 and @xmath188 , the hubble parameter and therefore @xmath177 are constant .
this means that @xmath189 as expected .
an important remark is that @xmath177 only depends on @xmath164 .
this means that , in the slow - roll approximation , @xmath177 only depends on the slow - roll parameter @xmath130 and not on the slow - roll parameter @xmath190 .
this is due to the fact that the calculation of @xmath177 only depends on background quantities .
the slow - roll parameter @xmath190 appears in the calculation through the generalized index @xmath169 , i.e. at the perturbed level .
we now discuss the spectrum obtained according to the initial state chosen . in this section
, we assume that the initial state is such that @xmath160 and @xmath161 . as a warm up ,
let us derive the spectrum in the particular case of de sitter , @xmath187 .
as mentioned above , this can only been done for gravitational waves since the de sitter limit is singular for density perturbations . for @xmath187
, we have @xmath191 and @xmath192 .
the spectrum can be determined exactly because the bessel functions in eqs .
( [ alpha ] ) and ( [ beta ] ) reduce to ordinary functions in the case @xmath187 .
the result reads @xmath193\ , , \\
\label{betds } \beta _ { _ { \rm t}}(k ) & = & \frac{1}{2}{\rm e}^{i\pi /2+i/\sigma _ 0 } ( 1 - 2\sigma _ 0 ^ 2)^{-1/4}\biggl[-i+\sigma _ 0+i\sigma _ 0 ^ 2+(i-\sigma _ 0 ) \sqrt{1 - 2\sigma _ 0 ^ 2}\biggr]\ , .
\end{aligned}\ ] ] the modulus of the two previous expressions can be easily derived from the above relation @xmath194 which coincides with the result of ref .
one can check that @xmath195 .
in addition , one notices that the result is independent on @xmath13 as expected for the de sitter space - time .
then , the power spectrum on super - horizon scales is given by @xmath196 where the super - horizon standard mode function can be expressed as @xmath197 .
the result reads @xmath198\ , .\ ] ] this spectrum is represented in fig .
[ tpl1 ] .
so far , no approximation has been made .
if we now assume that @xmath181 is a small quantity , we can expand the result in terms of this quantity .
let us also remark that the terms @xmath199 and @xmath200 are non - analytic for small values of @xmath181 . at leading order , we obtain that @xmath201 and @xmath202 in agreement with ref .
this leads to the following spectrum @xmath203\ , .\ ] ] therefore , the corresponding effect in the power spectrum is of order @xmath10 , i.e. a tiny effect if @xmath2 is small .
note , however , that it is not possible to distinguish this effect from a change in the parameter @xmath3 of the inflationary background .
we now turn to the case of slow - roll inflation .
we start with density perturbations .
the first step consists in calculating the coefficients @xmath204 and @xmath205 . at first order in the slow - roll parameters and at leading order in the parameter @xmath2
we have @xmath206 it is interesting to calculate the modulus of the coefficient @xmath207 .
straightforward calculations lead to @xmath208 we see that this quantity depends on both slow - roll parameters @xmath130 and @xmath190 , contrary to what was obtained in ref .
@xcite where the coefficient @xmath205 was found not to depend on the slow - roll parameter @xmath190 .
the main reason for this discrepancy is that , in ref .
@xcite , the spectrum of a scalar field on an unperturbed background cosmological model ( or , equivalently , the spectrum of gravitational waves ) was calculated , and not the spectrum of scalar metric fluctuations , see the discussion after eq .
( [ specscaf ] ) .
if one uses the expression of the scale factor @xmath209 and inserts this into the expression for the power spectrum for a scalar field on an unperturbed background , one does not obtain the correct expression for the power spectrum of density perturbations and one misses the terms proportional to the slow - roll parameter @xmath190 .
one only obtains either the power spectrum of gravitational waves or the power spectrum of density perturbations in the very particular case of power - law inflation for which we have @xmath210 ( and for which , in fact , the exact solution is known ) .
this is why putting @xmath211 in eq .
( [ modbeta ] ) reproduces the result found in ref .
@xcite .
then , the power spectrum of the conserved quantity @xmath63 is given by the following expression [ to be compared with ( [ srdp ] ) ] @xmath212 \nonumber \\ & & -\biggl[\frac32 \delta + 3\epsilon \ln \frac{k}{a_0m_{_{\rm c } } } + 2(c+1)\epsilon + 2c(\epsilon -\delta ) +
2(2\epsilon -\delta ) \ln \frac{k}{k_*}\biggr]\sin \biggl[\frac{2}{\sigma _ 0 } \biggl(1+\epsilon + \epsilon \ln \frac{k}{a_0m_{_{\rm c}}}\biggr)\biggr ] \nonumber \\ & & -\pi ( 2\epsilon -\delta ) \cos \biggl[\frac{2}{\sigma _ 0 } \biggl(1+\epsilon + \epsilon \ln \frac{k}{a_0m_{_{\rm c}}}\biggr ) \biggr]\biggr\ } \biggr)\ , .
\end{aligned}\ ] ] thus , the power spectrum has oscillations superimposed on the scale - invariant base . hence , in this case the change in the spectrum is `` in principle '' physically measurable . however , the prospects for actually detecting the difference of order @xmath213 between ( [ eq54 ] ) and ( [ srdp ] ) are hopeless if the parameter @xmath2 is small . indeed
if , for instance , we assume that @xmath214 , a value consistent with string theory , and if we consider that @xmath215 then the correction to the power spectrum is of order @xmath216 .
there is no possibility to detect such a small effect even with the planck satellite . for completeness ,
let us mention the slow - roll result for gravitational waves . in this case
the coefficients @xmath217 and @xmath218 can be deduced from @xmath204 and @xmath207 by putting @xmath210 .
then , the power spectrum is given by @xmath219 -\biggl[\frac32 \epsilon + 3\epsilon \ln \frac{k}{a_0m_{_{\rm c } } } + 2(c+1)\epsilon \nonumber \\ & & + 2\epsilon \ln \frac{k}{k_*}\biggr]\sin \biggl[\frac{2}{\sigma _ 0 } \biggl(1+\epsilon + \epsilon \ln \frac{k}{a_0m_{_{\rm c}}}\biggr)\biggr ] -\pi\epsilon \cos \biggl[\frac{2}{\sigma _ 0 } \biggl(1+\epsilon + \epsilon \ln \frac{k}{a_0m_{_{\rm c}}}\biggr ) \biggr]\biggr\ } \biggr)\ , .
\end{aligned}\ ] ] we can also estimate how the consistency check of inflation is modified . at first order in the slow - roll parameters and at leading order in the parameter @xmath2
, this quantity is the same as in the standard case , namely @xmath220 this is because at zeroth order in the slow - roll parameter , the leading contribution in @xmath2 is the same for gravitational waves and density perturbations .
we have established that , in the case of the instantaneous minkowski state , the correction to the power spectra is of order @xmath221 .
it is interesting to see whether this conclusion remains true for other initial states .
for this reason , we now repeat the calculation of the bogoliubov coefficients in the case of a de sitter space - time with an arbitrary @xmath222 -vacuum state , characterized by a value of @xmath157 and @xmath158 , as the initial state .
one finds @xmath223 \ , , \\
\beta _ { _ { \rm t}}(k ) & = & \frac{1}{2}(c_k+d_k){\rm e}^{i\pi/2 + i/\sigma _ 0 } ( 1 - 2\sigma _ 0 ^ 2)^{-1/4}\biggl[-i+\sigma _ 0+i\sigma _ 0 ^ 2+z_k(i-\sigma _ 0 ) \sqrt{1 - 2\sigma _ 0 ^ 2}\biggr ] \ , , \end{aligned}\ ] ] where we have introduced the quantity @xmath224 defined by @xmath225 . on
can easily check that for @xmath160 and @xmath161 , i.e. @xmath226 , the above expressions reduce to eqs .
( [ altds ] ) and ( [ betds ] ) .
one can also calculate the modulus of these bogoliubov coefficients , @xmath227\biggr\}\ , , \\
\vert \beta _ { _ { \rm t}}\vert ^2 & = & \frac{\vert c_k+d_k\vert ^2}{4 } \biggl\{\frac{(1+z_kz_k^*)-\sigma _ 0 ^ 2(1+z_kz_k^ * ) + \sigma _ 0 ^ 4(1 - 2z_kz_k^*)}{\sqrt{1 - 2\sigma _ 0 ^ 2 } } -[z_k^*+z_k+i(z_k^*-z_k)\sigma _ 0 ^ 3]\biggr\}\ , , \end{aligned}\ ] ] and verify that @xmath228 .
from the two previous relations , one can determine the power spectrum exactly .
the result reads @xmath229 \nonumber \\ & & + \frac{1}{2}(1-z_kz_k^*)(3\sigma _ 0 ^ 2 - 1 ) \cos \biggl(\frac{2}{\sigma _ 0}\biggr ) + ( 1-z_kz_k^*)\sigma _ 0\sin \biggl(\frac{2}{\sigma _ 0}\biggr ) -\frac{i}{2}(z_k^*-z_k)(1 - 2\sigma _ 0 ^ 2)^{3/2 } \sin \biggl(\frac{2}{\sigma _ 0}\biggr ) \nonumber \\ & & -\frac{i}{2}\sigma _ 0 ^ 3(z_k^*-z_k)\sqrt{1 - 2\sigma _ 0 ^ 2 } + \frac{i}{2}\sigma _ 0(z_k^*-z_k)(\sigma _ 0 ^ 2 - 2 ) \sqrt{1 - 2\sigma _ 0 ^ 2 } \cos \biggl(\frac{2}{\sigma _ 0}\biggr)\biggr\}\ , .
\end{aligned}\ ] ] as is apparent from the above expression , the correction in the power spectrum compared to the usual results is of order unity , i.e. not suppressed by any power of @xmath230 .
if one argues that the bogoliubov coefficients themselves should be of linear order in @xmath230 then the correction terms would also be linear in @xmath230 . on the other hand , unless the bogoliubov coefficients depend on @xmath13 , the effect is simply a change in the amplitude of the spectrum , and can hence be absorbed in a redefinition of the background .
thus , the effect is not physically measurable . on the other hand ,
if the coefficients depend on @xmath13 ( as discussed e.g. in the analysis of @xcite ) there is a large measurable effect on the power spectrum . for the instantaneous minkowski state ,
one check easily that only the two first terms in the curly brackets survives and that the corresponding expression reduces to the one found previously .
if we expand this spectrum in powers of @xmath231 , we find that @xmath232\sigma _ 0 + \cdots \biggr\}\ , .\ ] ] we conclude that the result obtained in the previous subsection for the instantaneous minkowske vacuum , namely that the corrections to the spectrum are suppressed by three powers of @xmath230 appears to be very particular to the choice of that state .
however , the previous analysis does not cover all the possible cases .
indeed , as we are going to demonstrate , the proposal put forward in ref .
@xcite corresponds in fact to a case where the complex number @xmath224 can also depend on @xmath2 .
the previous study assumed that @xmath157 and @xmath158 were pure complex numbers .
therefore , in danielsson s case @xcite , the analysis needs to be redone .
we start our analysis with the simplest case , namely gravitational waves in a de sitter background .
we treat this case in some detail since then these calculations can be used to study more complicated situations , for instance slow - roll inflation .
the einstein - hilbert action @xmath233 expanded to second order ( since we are dealing with first order equations of motion ) reads @xmath234a^2(\eta ) { \rm d}^4x\ , .
\nonumber \end{aligned}\ ] ] in the first of the two previous expressions , @xmath235 is the inverse of the flrw metric .
the next step consists in inserting the expression for @xmath19 , see eq .
( [ eq2 ] ) , into the action .
this gives @xmath236 where the fields @xmath237 , with @xmath238 , are defined by the following expression @xmath239 therefore , the action of gravitational waves is equivalent to the action of two decoupled scalar fields . inserting the fourier decomposition of the two scalar fields into the expression for the action ,
one arrives at @xmath240 + \biggl(\frac{a'{}^2}{a^2}-k^2\biggr ) \mu ^s_{_{\rm t}}(\mu ^s_{_{\rm t}})^*\biggr\}\ , .
\end{aligned}\ ] ] we can check that this lagrangian leads to the correct equation of motion : we have ( we did not take into account the overall constant ; here the bar over @xmath241 means that we are considering the lagrangian in the fourier space ) @xmath242 therefore , the euler - lagrange equations given by the well - known expression @xmath243
-\frac{\delta \bar{\cal l}}{\delta \mu _ { _ { \rm t}}^s}=0\ , , \ ] ] reproduce the correct equations of motion for the variables
let us notice that we can also vary the lagrangian with respect to @xmath245 to obtain the equation of motion for @xmath246 . from the lagrangian formalism that we have just described
, we can now pass to the hamiltonian formalism .
the conjugate momentum to @xmath247 is defined by the formula @xmath248\ , .\ ] ] in real space , the conjugate momentum @xmath249 is defined by the expression @xmath250 we can check that the two definitions are consistent by means of the relation @xmath251 danielsson s boundary condition consists in demanding that , at the time of creation @xmath151 , one has the usual relation characteristic of the standard vacuum state , namely @xmath252 using the fourier decomposition of the scalar fields @xmath253 and of their conjugate momenta , this last equation boils down to @xmath254 this relation is to be satisfied at the time @xmath255 , and implies a link between the coefficients @xmath217 and @xmath256 . since , in addition , @xmath257 , the coefficients are in fact completely fixed and they now depend on the parameter @xmath2 .
explicitly , the link can be expressed as @xmath258 so far no approximation has been made and the previous relation is general .
we now restrict our study to the de sitter case . using the explicit form of the mode function in this case ,
one arrives at @xmath259 which is exactly the relation found in ref .
@xcite . from the normalization
, we deduce that @xmath260 .
the exact power spectrum can now be determined since we explicitly know the coefficients @xmath217 and @xmath261 . performing the standard calculation
, one finds @xmath262\ , .\ ] ] this expression should be compared with the corresponding equation found in the case of the minkowski instantaneous state , see eq .
( [ minkops ] ) .
in particular , expanding everything in terms of @xmath2 , one obtains @xmath263\ , , \ ] ] i.e. , a first order effect instead of a third order effect . however , once again note that this effect can be reproduced by redefining the background cosmological parameters , and hence it is not a physically measurable effect .
the spectrum is represented in fig .
[ tpl2 ] .
let us now turn to the case of slow - roll inflation .
the goal is simply to evaluate the ratio @xmath264 given by eq .
( [ ratioab ] ) to leading order in the parameter @xmath2 and to first order in the slow - roll parameter @xmath119 .
one finds @xmath265 of course , for @xmath188 , one recover the exact result obtained previously at leading order in @xmath2 .
finally the power spectrum for gravitational waves reads @xmath266 \nonumber \\ & & \times \sin \biggl[\frac{2}{\sigma _ 0 } \biggl(1+\epsilon + \epsilon \ln \frac{k}{a_0m_{_{\rm c}}}\biggr)\biggr ] + \sigma _ 0\pi \epsilon \cos \biggl[\frac{2}{\sigma _ 0 } \biggl(1+\epsilon + \epsilon \ln \frac{k}{a_0m_{_{\rm c}}}\biggr)\biggr]\biggr\}\ , .
\end{aligned}\ ] ] note that since in this case the correction terms lead to oscillations in the spectrum about a scale - invariant base spectrum , the effect of the correction terms is in principle measurable . since the amplitude is only suppressed by one power of @xmath2 , the prospects of being able to detect such effects in upcoming experiments are good .
we now explore the same mechanism but for scalar metric ( density ) perturbations .
the action is expressed in terms of the variable @xmath82 introduced before .
it reads @xmath267\ , , \ ] ] in fourier space , the action reads @xmath268 \mu _ { _ { { \rm s}}}\mu _ { _ { { \rm s}}}^*\biggr\}\ , .\ ] ] the conjugate momentum to the variable @xmath82 is given by @xmath269 and therefore danielsson s boundary condition reads @xmath270 where this relation must be evaluated at the time @xmath255 .
we note that the term @xmath271 , which was present in the case of the scalar field , does not appear for density perturbations .
this can be traced back to the fact that the two quantities that are quantized , @xmath34 and @xmath82 do not have the same action .
this means that the link between @xmath272 and @xmath205 for density perturbations is not the same as for gravitational waves . for scalar metric fluctuations
, it reads @xmath273 this equation should be compared with eq .
( [ ratioab ] ) .
the absence of the terms @xmath274 has important consequences . in order to guess what the difference is
, we can apply the previous equation to the de sitter case , even if in principle this is not allowed ( see the discussion above ) .
one finds @xmath275 expanding the previous expression in @xmath2 , we find that the first order terms cancel out and we are left with @xmath276 .
therefore , the result will be of order @xmath11 and not of order @xmath2 as it was the case for a scalar field and gravitational waves .
we would have obtained a linear correction in the power spectrum if danielsson s condition had been @xmath277 as it is for gravitational waves , instead of @xmath278 , see eq .
( [ condscalar ] ) .
let us now evaluate eq .
( [ bascalar ] ) consistently and rigorously in the slow - roll approximation .
one finds @xmath279 as expected , the result is quadratic in @xmath2 . repeating the standard calculations , one can find the explicit expression for the power spectrum @xmath280\cos \biggl[\frac{2}{\sigma _ 0 } \biggl(1+\epsilon + \epsilon \ln \frac{k}{a_0m_{_{\rm c}}}\biggr)\biggr ] + \sigma _ 0 ^ 2\pi ( 2\epsilon -\delta ) \sin \biggl[\frac{2}{\sigma _ 0 } \biggl(1+\epsilon + \epsilon \ln \frac{k}{a_0m_{_{\rm c}}}\biggr)\biggr ] \biggr\ } \ , .
\nonumber \\ \end{aligned}\ ] ] as in the case of gravitational waves , in principle this is a measurable effect since it corresponds to oscillations of the power spectrum .
typical power spectra are represented in fig .
since the correction is quadratic in @xmath2 , the correction turns out to be extremely small and the prospects for the detection of such an effect are not optimistic even with a high - accuracy experiment like planck . thus , at least with this choice of the precise form of the action @xcite , danielsson s prescription leads to trans - planckian corrections of different strengths for gravitational waves and for scalar fluctuations .
note that this also implies that the consistency relation of inflation will be modified . comparing ( [ eq78 ] ) and ( [ pssrs2 ] ) , at leading order in @xmath2 ,
one finds @xmath281\biggr\}\ , , \ ] ] where the ratio has been evaluated at the pivot scale .
as expected from the previous considerations , the correction is linear in @xmath181 .
in this section , we consider a case where the trans - planckian physics is described by mean of a modified dispersion relation .
although , this phenomenological description is different from the one considered above , it has already been shown that the corresponding spectrum can also possess superimposed oscillations @xcite . in this section ,
we calculate the spectrum of a scalar field ( or gravitational waves ) living in a de sitter space - time and we compare this result with the corresponding result obtained in the framework described in the previous sections , see eqs .
( [ minkops ] ) and ( [ dpsgw ] ) .
in particular , our goal is to determine which parameter controls the magnitude of the correction to the standard power spectrum in the case of a modified dispersion relation ( for the case treated before , it was the parameter @xmath2 ) .
the main shortcomings of describing the trans - planckian physics by a modified dispersion relation is that we need to assume something about the physics beyond the planck , contrary to the kind of modification envisaged above .
modifications in the power spectrum are obtained if the wkb approximation is violated in the trans - planckian regime . a typical example where this happens is for the dispersion relation @xmath282 , which represents the first terms of a systematic taylor expansion .
this dispersion relation is represented in fig .
[ disp ] .
the equation of motion for the mode function of a scalar field is now given by @xmath283\mu=0\ , , \ ] ] where @xmath65 is the comoving frequency given by @xmath284 $ ] .
the difference compared to the previous section is that the equation determining the behavior of the mode function is modified .
the mode functions depend on whether @xmath285 or @xmath286 .
let us consider an expanding cosmological background .
there will be four time intervals ( `` regions '' ) , see fig .
[ piece ] . as the universe expands , the physical wavenumber will decrease .
the evolution starts in region i , where the wkb approximation is valid , the term @xmath287 dominates and the mode function is given by @xmath288 we have chosen the wkb vacuum , i.e. we have retained only one branch of the most general solution . at a time @xmath289 , the mode enters region ii , where the wkb approximation is violated , and the mode function can be expressed as @xmath290 the coefficients @xmath291 can be determined from joining the mode function and its derivative at the matching point @xmath292 .
after more expansion [ at the time @xmath293 , the mode leaves region ii and enters region iii , where the wkb approximation is restored .
however , now the mode function has two branches , the second branch having been `` generated '' in region ii .
it is here that the trans - planckian physics modifies the standard result ( which predicts @xmath294 ) .
the mode function can be written as @xmath295 the coefficients @xmath296 and @xmath297 can be found in terms of @xmath298 by matching the solution ( and its derivative ) at @xmath299 as previously .
finally , in region iv , the wkb approximation is again violated .
this region corresponds to the usual super - hubble region .
we have @xmath300 the transition between regions iii and iv occurs at the time @xmath301 , the time of hubble radius crossing .
our aim is to calculate the coefficients @xmath302 which determines the spectrum of the growing mode . performing the matching at @xmath303
, we obtain @xmath304 where the subscript `` @xmath305 '' means that the corresponding quantity is evaluated at @xmath303 . in the previous equation ,
we have introduced the notation @xmath306 and @xmath307 . on the other hand
, a lengthy but straightforward calculation shows that the coefficients @xmath296 and @xmath297 are given by @xmath308{\rm e}^{i(\omega _ 2-\omega _ 1)}\ , , \\
\beta _ k & = & -\frac{i}{\sqrt{4\omega _ 1\omega _ 2 } } \biggl[\frac{a_2}{a_1}\gamma
_ 2-\frac{a_1}{a_2}\gamma _ 1 -a_1a_2\gamma _ 1\gamma _ 2\int _
{ \eta _ 1}^{\eta _ 2}\frac{{\rm d}\tau } { a^2(\tau ) } \biggr]{\rm e}^{-i(\omega _ 2+\omega _ 1)}\ , .
\end{aligned}\ ] ] these expressions are similar to those found in refs .
@xcite and @xcite .
so far , no approximation has been made .
to go further , one has to take into account the fact that the difference @xmath309 can not be too large .
otherwise , this would mean that particles production in region iii is too important and , as a consequence , that the calculation is not valid due to this back - reaction problem .
therefore , we can write @xmath310 and perform an expansion of the coefficients @xmath311 and @xmath297 in terms of the parameter @xmath312 . the result reads @xmath313+{\cal o}(\delta^2 ) \
, , \quad \beta _ k=\frac{i}{2}{\rm e}^{-i(\omega _ 2+\omega _ 1 ) } \omega _ 2\eta _ 2\biggl(1-\frac{q_2}{\omega _ 2 ^ 2 } + \frac{a_2''}{\omega _ 2 ^ 2a_2}\biggl)\delta + { \cal o}(\delta^2 ) \ , .\ ] ] in this expression , @xmath314 is the parameter which controls the accuracy of the wkb approximation .
it is defined by @xmath315 @xcite .
as expected , if we send the parameter @xmath312 to zero , then @xmath297 vanishes and @xmath296 becomes unity . inserting the above expressions for @xmath296 and @xmath297 into eq .
( [ mu4 ] ) , one arrives at @xmath316\delta + { \cal o}(\delta ^2)\biggr\}\ , .\ ] ] notice that in order to establish the previous expression we have just assumed that @xmath312 is a small number but we have not assumed that @xmath317 or @xmath318 are small . as expected , the correction to @xmath319 is given by the bogoluibov coefficient @xmath320 .
we now need to calculate explicitly the previous quantities . for this purpose
, we introduce an approximate piecewise form of the dispersion relation considered before , namely @xmath321 this piecewise dispersion relation is represented in fig .
[ piece ] .
the parameter @xmath222 controls the slope of the dispersion relation in the region where the group velocity does not have the same sign as the phase velocity .
the parameter @xmath322 controls the slope in the region @xmath323 .
assuming that the space - time is de sitter , one can also compute the comoving frequency @xmath324 @xmath325}\ , & $ k_{_{\rm phys } } > k _ * \ , .$ } \ ] ] obviously , one has to choose @xmath326 in order to insure that the frequency remains positive .
the various times of matching can now be calculated very simply .
they read @xmath327\ , , \\
\eta _ 2(k ) & = & -\frac{1}{k}\biggl(-\frac{\sqrt{2}}{\alpha } + \frac{\alpha + 1}{\alpha } \frac{k_{_{\rm c}}}{h}\biggr)\ , , \\ \eta _ 3(k ) & = & -\frac{1}{k}\ , .
\end{aligned}\ ] ] the times at which @xmath328 and @xmath329 , i.e. @xmath330 and @xmath331 respectively ( see fig . [ piece ] ) , can also be determined easily .
they are given by @xmath332 and @xmath333 .
we have based the calculation of @xmath319 on the assumption that the parameter @xmath312 is small in order to avoid a back - reaction problem .
this means that the time spent by the modes of interest in the region where the wkb approximation is violated is small . in turn
, this requires a link between the two scales @xmath334 and @xmath148 which characterizes the shape of the dispersion relation and the hubble constant @xmath26 which characterizes the `` velocity '' with which a mode crosses the region where the wkb approximation is not valid .
using the expressions of @xmath292 and @xmath335 , one finds that the link between @xmath148 , @xmath336 and @xmath312 can be expressed as @xmath337 \biggl[-\sqrt{2}+(\alpha + 1)\frac{k_{_{\rm c}}}{h}\biggr]\ , .\ ] ] if @xmath338 and @xmath339 , then one has @xmath340 , as expected .
the final result can be expressed in terms of @xmath317 and @xmath312 only .
we can now calculate each term present in eq .
( [ d+ ] ) .
straightforward calculations show that @xmath341 , @xmath342 and @xmath343 . in order to calculate the integral appearing in the argument of the sine function in eq .
( [ d+ ] ) , one has to cut it into several pieces and to use the corresponding form of the piecewise dispersion relation .
one obtains @xmath344 \nonumber \\ & & -\sqrt{2}\delta \ , .
\end{aligned}\ ] ] we have to insert this last expression into the sine function and expand the resulting expression once more in terms of the parameter @xmath345 .
the final result reads @xmath346\sin \biggl[(\sqrt{2}-1 ) -(\alpha + 1)\frac{k_{_{\rm c}}}{h } \ln \biggl(\frac{1}{\alpha + 1-\sqrt{2}h / k_{_{\rm c}}}\biggr ) \biggr]\delta \nonumber \\ & & + { \cal o}(\delta ^2 ) \biggr\}\ , .
\end{aligned}\ ] ] as expected the result does not depend on @xmath13 since the scalar field lives in de sitter space - time .
it is worth noticing that we have not assumed anything about the ratio @xmath317 in order to obtain the previous expression ( only the parameter @xmath312 was supposed to be small ) .
the previous expression is the main result of this section .
it should be compared with eqs .
( [ minkops ] ) and ( [ dpsgw ] ) obtained previously .
we see that the magnitude of the correction is no longer controled by any power of the ratio of two scales as it was before .
the magnitude of the effect is determined by the time spent in the region where the wkb approximation is violated which in turn depends on the shape of the dispersion relation that was assumed .
finally , let us return to the back - reaction problem .
there is no back - reaction problem if @xmath347 .
using the expression of @xmath297 derived before and assuming that @xmath348 , we see that the parameter @xmath312 must satisfy @xmath349 .
this amounts to a severe fine - tuning of the scales @xmath336 , @xmath148 and of the hubble parameter @xmath26 during inflation in order to satisfy this condition .
we have analyzed the magnitude of correction terms to the power spectra of scalar metric ( density ) fluctuations and gravitational waves in inflationary cosmology under the assumption that fluctuation modes are generated when their physical length scale equals some critical length determined by the unknown planck - scale physics , but without modifying the equations of motion for the fluctuations .
the magnitude of the correction terms can then be expressed as a function of the dimensionless ratio @xmath350 , where @xmath351 is the characteristic hubble expansion rate during inflation , and @xmath1 is the mass scale at which the new physics sets in . it is important to realize that the magnitude of the correction terms is in general different for gravitational waves and for scalar metric fluctuations - a point not realized in some papers on the `` trans - planckian problem '' of inflationary cosmology .
in addition , the magnitude of the correction terms depends sensitively on the initial state chosen .
we have shown that for the local minkowski vacuum state , the correction terms are of the order @xmath352 ( in agreement with the results of @xcite ) , whereas for nontrivial @xmath9-vacua the effects are much larger . if the bogoliubov coefficients which describe the mode mixing do not depend on @xmath230 , then the correction terms can be of order unity . in the case of danielsson s @xmath9-vacuum
for which the bogoliubov coefficients depend on @xmath230 , the corrections to the gravitational wave spectrum are suppressed by one power of @xmath230 ( in agreement with the results of @xcite ) , the corrections to the scalar metric fluctuation spectrum by two powers of @xmath230 .
thus , the consistency relation for fluctuations in inflationary cosmology obtains corrections of linear order in @xmath230 , as already emphasized in @xcite . in the final section of the paper we compared the results obtained in earlier sections with the results obtained by assuming that trans - planckian physics leads to a modified dispersion relation . in this case , corrections to the usual power spectra of fluctuations can be obtained which are not suppressed by any small dimensionless combinations of energy scales in the problem . however , demanding that the back - reaction remains under control leads to severe fine - tuning requirements on such models .
r. h. brandenberger , proceeding of the international school on cosmology , kish island , iran ( 2000 ) , [ arxiv : hep - ph/9910410 ] .
j. martin and r. h. brandenberger , phys .
d * 63 * , 123501 ( 2001 ) , [ arxiv : hep - th/0005209 ] .
r. h. brandenberger and j. martin , mod .
a * 16 * , 999 ( 2001 ) , [ arxiv : astro - ph/0005432 ] . w. unruh , phys . rev . *
d51 * , 2827 ( 1995 ) , [ arxiv : gr - qc/9409008 ] . s. corley and t. jacobson , phys .
* d54 * , 1568 ( 1996 ) , [ arxiv : hep - th/9601073 ] ; s. corley , phys . rev . *
d57 * , 6280 ( 1998 ) , [ arxiv : hep - th/9710075 ] . j. c. niemeyer , phys . rev . *
d63 * , 123502 ( 2001 ) , [ arxiv : astro - ph/0005533 ] ; j. c. niemeyer , [ arxiv : astro - ph/0201511 ] . j. martin and r. h. brandenberger , proceedings of the ninth marcel grossmann meeting on general relativity , edited by r. t. jantzen , v. gurzadyan and r. ruffini , world scientific , singapore , 2002 , [ arxiv : astro - ph/0012031 ] .
j. c. niemeyer and r. parentani , phys . rev .
* d64 * , 101301 ( 2001 ) , [ arxiv : astro - ph/0101451 ] .
m. lemoine , m. lubo , j. martin and j. p. uzan , phys .
rev . * d65 * , 023510 ( 2002 ) , [ arxiv : hep - th/0109128 ] .
r. h. brandenberger , s. e. joras and j. martin , phys .
d * 66 * , 083514 ( 2002 ) , [ arxiv : hep - th/0112122 ]
. s. shankaranarayan , class .
. grav . * 20 * , 75 ( 2003 ) , [ arxiv : gr - qc/0203060 ] .
a. kempf , phys .
* d63 * ( 2001 ) 083514 , [ arxiv : astro - ph/0009209 ] . c. s. chu , b. r. greene and g. shiu , mod .
a16 * , 2231 ( 2001 ) , [ arxiv : hep - th/0011241 ] .
r. easther , b. r. greene , w. h. kinney and g. shiu , phys . rev . *
d64 * , 103502 ( 2001 ) , [ arxiv : hep - th/0104102 ] ; r. easther , b. r. greene , w. h. kinney and g. shiu , [ arxiv : hep - th/0110226 ] .
a. kempf and j. c. niemeyer , phys .
* d64 * , 103501 ( 2001 ) , [ arxiv : astro - ph/0103225 ] . l. hui and w. h. kinney , [ arxiv : astro - ph/0109107 ] .
r. brandenberger and p. m. ho , phys .
d * 66 * , 023517 ( 2002 ) , [ aapps bull . * 12n1 * , 10 ( 2002 ) ] , [ arxiv : hep - th/0203119 ] . s. f. hassan and m. s. sloth , [ arxiv : hep - th/0204110 ] .
f. lizzi , g. mangano , g. miele and m. peloso , noncommutative geometry , jhep * 0206 * , 049 ( 2002 ) , [ arxiv : hep - th/0203099 ] .
u. h. danielsson , phys .
d * 66 * , 023511 ( 2002 ) , [ arxiv : hep - th/0203198 ] ; u. h. danielsson , jhep * 0207 * , 040 ( 2002 ) , [ arxiv : hep - th/0205227 ] .
k. goldstein and d. a. lowe , phys .
d * 67 * , 063502 ( 2003 ) , [ arxiv : hep - th/0208167 ] . g. l. alberghi , r. casadio and a. tronconi , [ arxiv : gr - qc/0303035 ] . l. bergstrom and u. h. danielsson , jhep * 0212 * , 038 ( 2002 ) , [ arxiv : hep - th/0211006 ] . t. banks and l. mannelli , [ arxiv : hep - th/0209113 ] . m. b. einhorn and f. larsen , phys . rev .
d * 67 * , 024001 ( 2003 ) , [ arxiv : hep - th/0209159 ] .
n. kaloper , m. kleban , a. lawrence , s. shenker and l. susskind , jhep * 0211 * , 037 ( 2002 ) , [ arxiv : hep - th/0209231 ] .
k. goldstein and d. a. lowe , [ arxiv : hep - th/0302050 ] .
r. easther , b. r. greene , w. h. kinney and g. shiu , phys .
d * 66 * , 023518 ( 2002 ) , [ arxiv : hep - th/0204129 ] . j. c. niemeyer , r. parentani and d. campo , phys .
d * 66 * , 083510 ( 2002 ) , [ arxiv : hep - th/0206149 ] . n. kaloper , m. kleban , a. e. lawrence and s. shenker , phys . rev .
d * 66 * , 123510 ( 2002 ) , [ arxiv : hep - th/0201158 ] .
r. h. brandenberger and j. martin , int .
j. mod .
a * 17 * , 3663 ( 2002 ) , [ arxiv : hep - th/0202142 ] .
c. p. burgess , j. m. cline , f. lemieux and r. holman , jhep * 0302 * , 048 ( 2003 ) , [ arxiv : hep - th/0210233 ] .
v. bozza , m. giovannini and g. veneziano , [ arxiv : hep - th/0302184 ] . c. armendariz - picon and e. a. lim , [ arxiv : hep - th/0303103 ] .
d. j. chung , a. notari and a. riotto , [ arxiv : hep - ph/0305074 ] . v. f. mukhanov , h. a. feldman and r. h. brandenberger , phys .
* 215 * , 203 ( 1992 ) .
l. p. grishchuk , zh .
* 64 * , 825 ( 1974 ) , [ sov
jetp * 40 * , 409 ( 1974 ) ] . j. martin and d. j. schwarz , phys . rev .
d * 67 * , 083512 ( 2003 ) , [ arxiv : astro - ph/0210090 ] . j. m. bardeen , phys .
d * 22 * , 1882 ( 1980 ) .
v. f. mukhanov , sov .
jetp * 67 * , 1297 ( 1988 ) [ zh .
fiz . * 94n7 * , 1 ( 1988 zetfa,94,1 - 11.1988 ) ]
. j. martin and d. j. schwarz , phys .
d * 57 * , 3302 ( 1998 ) , [ arxiv : gr - qc/9704049 ]
. j. e. lidsey , a. r. liddle , e. w. kolb , e. j. copeland , t. barreiro and m. abney , rev .
phys . * 69 * , 373 ( 1997 ) , [ arxiv : astro - ph/9508078 ] .
d. j. schwarz , c. a. terrero - escalante and a. a. garcia , phys .
b * 517 * , 243 ( 2001 ) , [ arxiv : astro - ph/0106020 ] .
j. martin and d. j. schwarz , phys .
d * 62 * , 103520 ( 2000 ) , [ arxiv : astro - ph/9911225 ] .
j. martin , a. riazuelo and d. j. schwarz , astrophys .
j. * 543 * , l99 ( 2000 ) , [ arxiv : astro - ph/0006392 ] .
ulf danielsson , private communication . | we calculate the power spectrum of metric fluctuations in inflationary cosmology starting with initial conditions which are imposed mode by mode when the wavelength equals some critical length @xmath0 corresponding to a new energy scale @xmath1 at which trans - planckian physics becomes important . in this case
, the power spectrum can differ from what is calculated in the usual framework ( which amounts to choosing the adiabatic vacuum state ) .
the fractional difference in the results depends on the ratio @xmath2 between the hubble expansion rate @xmath3 during inflation and the new energy scale @xmath1 .
we show how and why different choices of the initial vacuum state ( stemming from different assumptions about trans - planckian physics ) lead to fractional differences which depend on different powers of @xmath2 . as we emphasize
, the power in general also depends on whether one is calculating the power spectrum of density fluctuations or of gravitational waves . |
at the turn of the last century observations from ia supernova and cosmic microwave background ( cmb)radiation confirmed that our universe is suffering from an accelerated expansion @xcite , but the physical origin of this acceleration is yet to be known .
the standard explanation invokes an unknown `` dark energy '' component which has the property that positive energy density and negative pressure .
observations indicate that dark energy occupies about 70% of the total energy of the universe , and the contribution of dark matter is @xmath0 26% .
this accelerated expansion of the universe has also been strongly confirmed by some other independent experiments like sloan digital sky survey ( sdss ) @xcite , baryonic acoustic oscillation ( bao ) @xcite , wmap data analysis @xcite etc . over the past decade
there have been many theoretical models for mimicking the dark energy behaviors , such as the simplest ( just ) cosmological constant in which the equation of state is independent of the cosmic time and which can fit the observations well .
this model is the so - called @xmath1cdm , containing a mixture of cosmological constant @xmath1 and cold dark matter ( cdm ) .
however , two problems arise from this scenario , namely `` fine - tuning '' and the `` cosmic coincidence '' problems . in order to solve these two problems ,
many dynamical dark energy models were suggested , whose equation of state evolves with cosmic time . the scalar field or quintessence
@xcite is one of the most favored candidate of dark energy which produce sufficient negative pressure to drive acceleration . in order to alleviate the cosmological - constant problems and explain the acceleration expansion ,
many dynamical dark energy models have been proposed , such as k - essence , tachyon , phantom , quintom , chaplygin gas model , etc @xcite . also the interacting dark energy models including modified chaplygin gas @xcite , holographic dark energy model @xcite , and braneworld model @xcite have been proposed .
the equation of state of modified chaplygin gas is given by , @xmath2 where @xmath3 and @xmath4 are respectively the pressure and energy density and @xmath5 , @xmath6 and @xmath7 are positive constants . in einstein s gravity ,
the modified chaplygin gas @xcite best fits with the 3 year wmap and the sdss data with the choice of parameters @xmath8 and @xmath9 @xcite which are improved constraints than the previous ones @xmath10 @xcite . +
another possibility is that general relativity is only accurate on small scales and has to be modified on cosmological distances .
one of these is modified gravity theories . in this case cosmic acceleration
would arise not from dark energy as a substance but from the geometry of space - time i.e. from the dynamics of modified gravity .
modified gravity constitutes an interesting dynamical alternative to @xmath1cdm cosmology in that it is also able to describe the current cosmic acceleration .
the simplest modified gravity is dgp brane - world model @xcite .
the other alternative approach dealing with the acceleration problem of the universe is changing the gravity law through the modification of action of gravity by means of using @xmath11 gravity @xcite instead of the einstein - hilbert action .
some of these models , such as @xmath12 and logarithmic models , provide an acceleration for the universe at the present time @xcite .
other modified gravity includes @xmath13 gravity , @xmath14 gravity , gauss - bonnet gravity , horava - lifshitz gravity , brans - dicke gravity , etc @xcite . in recent times
there have been a lot of research on the background dynamics of different de models in modified gravity theories in the quest of a standard model of cosmology @xcite .
+ in the present work , we concentrate on the generalized einstein - aether ( ea ) theories as proposed by zlosnik et al @xcite , which is a generalization of the einstein - aether theory developed by jacobson et al @xcite . in recent years
a lot of work has been done in generalized einstein - aether theories @xcite . in the generalized einstein - aether theories by taking a special form of the lagrangian density of aether field
, the possibility of einstein - aether theory as an alternative to dark energy model is discussed in detail , that is , taking a special aether field as a dark energy candidate and it has been found the constraints from observational data @xcite . since modified gravity theory may be treated as alternative to dark energy , so meng et al @xcite have not taken by hand any types of dark energy in einstein - aether gravity and shown that the gravity may be generates dark energy . here
if we exempt this assumption , so we need to consider the dark energy from outside .
so we assume the frw universe in einstein - aether gravity model filled with the dark matter and the modified chaplygin gas ( mcg ) type dark energy .
this paper is organized as follows : section 2 comprises of the general concepts of einstein - aether gravity . in section 3
, we do an extensive study of the dynamical system . in section 4 ,
a detailed graphical analysis for the phase plane is presented .
finally the paper ends with some concluding remarks in section 5 .
in order to include lorentz symmetry violating terms in gravitation theories , apart from some noncommutative gravity models , one may consider existence of preferred frames .
this can be achieved admitting a unit timelike vector field in addition to the metric tensor of spacetime .
such a timelike vector implies a preferred direction at each point of spacetime . here the unit timelike vector field is called the _ aether _ and the theory coupling the metric and unit timelike vector is called the _ einstein - aether _ theory @xcite .
so einstein - aether theory is the extension of general relativity ( gr ) that incorporates a dynamical unit timelike vector field ( i.e. , aether ) . in the last decade
there is an increasing interest in the aether theory .
+ the action of the einstein - aether gravity theory with the normal einstein - hilbert part action can be written in the form @xcite @xmath15\ ] ] where @xmath16 is the vector field lagrangian density while @xmath17 denotes the lagrangian density for all other matter fields .
the lagrangian density for the vector part consists of terms quadratic in the field @xcite : @xmath18 @xmath19 @xmath20 where @xmath21 are dimensionless constants , @xmath22 is the coupling constant which has the dimension of mass , @xmath23 is a lagrange multiplier that enforces the unit constraint for the time - like vector field , @xmath24 is a contravariant vector , @xmath25 is metric tensor and @xmath26 ia an arbitrary function of @xmath27 . from ( 1 ) , we get the field equations @xmath28 @xmath29 where @xmath30 here @xmath31 is the energy momentum tensor for matter field and @xmath32 is the energy momentum tensor for the vector field and they are respectively given as follows : @xcite @xmath33 where @xmath4 and @xmath3 are respectively the energy density and pressure of matter and @xmath34 is the fluid 4-velocity vector and @xmath35-y_{(ab)}f'+\frac{1}{2}~g_{ab}m^{2}f+\lambda a_{a}a_{b}\ ] ] with @xmath36\ ] ] where the subscript @xmath37 means symmetric with respect to the indices involved and * the vector @xmath24 is considered to have time - like direction and satisfies @xmath38 to fix the variation of the action with respect to @xmath23 .
the normalized vector field @xmath24 has components @xmath39 in the frw cosmology with homogeneous and isotropic universe filled with perfect fluid . *
we consider the friedmann - robertson - walker ( frw ) metric of the universe as @xmath40\ ] ] where @xmath41 is the curvature scalar and @xmath42 is the scale factor . from equations ( 3 ) and ( 4 ) ,
we get @xmath43 where @xmath44 is constant . from eq . ( 5 ) , we get the modified friedmann equation for einstein - aether gravity as in the following @xcite : @xmath45 and @xmath46 where @xmath47 is hubble parameter .
now we see that if the first expressions of l.h.s . of equations
* ( 14 ) * and * ( 15 ) * are zero , we get the usual field equations for einstein s gravity . so first expressions arise for einstein - aether gravity .
also the conservation equation is given by @xmath48 now , assume that the matter fluid is combination of dark matter and modified chaplygin gas type dark energy .
so @xmath49 and @xmath50 , where @xmath51 and @xmath52 are respectively the energy density and pressure of dark matter and @xmath53 and @xmath54 are respectively the energy density and pressure of modified chaplygin gas .
assume that the dark matter follows the barotropic equation of state @xmath55 , where @xmath56 is a constant .
the equation of state of modified chaplygin gas ( mcg ) is given by @xcite @xmath57 where @xmath58 , @xmath59 and @xmath60 .
as in the present problem the interaction between de and pressureless dm has been taken into account for interacting de and dm the energy balance equation will be @xmath61 @xmath62 where @xmath63 is the interaction term , @xmath64 is the coupling parameter ( or transfer strength ) and @xmath65 is the total cosmic energy density which satisfies the energy conservation equation @xmath66 @xcite . since we lack information about the fact , how does de and dm interact so we are not able to estimate the interaction term from the first principles .
however , the negativity of @xmath67 immediately implies the possibility of having negative de in the early universe which is overruled by the necessity of the second law of thermodynamics to be held @xcite
. hence @xmath67 must be positive and small . from the observational data of 182 gold type ia supernova samples , cmb data from the three year wmap survey and the baryonic acoustic oscillations from the sloan digital sky survey , it is estimated that the coupling parameter between dm and de must be a small positive value ( of the order of unity ) , which satisfies the requirement for solving the cosmic coincidence problem and the second law of thermodynamics @xcite . due to the underlying interaction , the beginning of the accelerated expansion
is shifted to higher redshifts .
the continuity equations for dark energy and dark matter are given in equations ( 18 ) and ( 19 ) .
now we proceed to study the dynamical system . in this subsection
we plan to analyze the dynamical system .
for that firstly we convert the physical parameters into some dimensionless form , given by @xmath68 where the present value of the scale factor @xmath69 is assumed . * following the works of zlosnik et al and zuntz et al @xcite , we assume the ansatz for @xmath70 as * @xmath71 * where cosmic acceleration is realized for @xmath72 and for certain value of n @xmath73 without the need of a source term in the modified einstein equations @xcite , here modelled with the chaplygin gas . this form is efficient enough to express a wide range of behaviour .
following modified newtonian dynamics , if we use the same branch of @xmath70 as it uses on small scales , it can be shown that the background cosmology becomes inconsistent with the late time cosmic acceleration @xcite .
hence we use a reasonable form for @xmath70 given in eqn .
( 27 ) , which works for the accelerated regime ( @xmath74 ) . using this form of @xmath70 ,
the modified friedmann equations become , * @xmath75h^{2}=\frac{8\pi g}{3}\rho\ ] ] * where @xmath76 .
moreover the expression for @xmath77 is obtained as , * @xmath78 * where @xmath79 and @xmath80 is the present value of hubble constant*. * let us consider some special cases of @xmath81 .
we see that if @xmath82 , the friedmann equations are unchanged ( @xmath83 ) , and invariably there is no effect on the background cosmology .
if @xmath84 , the cosmological constant is recovered .
when @xmath85 , we have @xmath86 and there is a change of scale for the newton s constant by a factor of @xmath87 @xcite .
a schematic representation of the late time evolution of the universe depending on the value of @xmath81 can be found in @xcite . from the above discussion the most suitable choice of @xmath81 for our present assignment
is @xmath85 .
moreover it provides a bit of mathematical simplicity as well .
this value of @xmath81 completely lies in the accepted region which is @xmath88 .
as apparently conceived , this chosen value of @xmath81 does not actually hamper the generality of the problem .
in fact from our analysis we have seen that a slight variation in the value of @xmath81 does not bring about significant alterations to the final result . * using eqns .
( 13 - 19 ) into eqn.(20 ) we get the parameter gradients as @xmath89 @xmath90 and @xmath91 where @xmath92 is the eos parameter for mcg determined as @xmath93 @xmath94\ ] ] here @xmath95 is given by , @xmath96 the critical points of the above system are obtained by putting @xmath97 .
but due to the complexity of these equations , it is not possible to find a solution in terms of all the involved parameters .
* so we are compelled to search for a viable numerical solution . in order to accomplish this
, we have to replace the parameters by suitable numerical values which must be realistic cosmologically .
as far as the parameters of mcg are concerned , we will be using the values which are consistent with the observational data @xcite .
we consider @xmath98 .
the choice of the aether gravity parameters will not be so straightforward , since we do not have cosmologically constrained parametric values at our disposal . in @xcite
clifton et al has shown that in order to realize a more exotic but viable scenario , one should have @xmath99 . from eqn.(23 ) , we get an idea of the value of @xmath100 . if we take @xmath101 ( acceptable value from the latest observational data ) and @xmath85 , we see that in order to make @xmath102 , @xmath100 should be restricted to negative values .
although we get an idea about the sign of @xmath100 yet there is no information available in literature about the magnitude .
so keeping in mind the simplicity of calculations , we choose @xmath103 . using the accepted values of parameters in eqn.(23 )
we obtain @xmath104 approximately .
since @xmath22 has the dimensions of mass , so it is obvious that it should have large values .
the exact values of @xmath22 and @xmath64 are obtained from numerical fine tuning after putting the accepted numerical values of other parameters .
we take , @xmath105 and obtain the following critical point , * @xmath106 the critical point correspond to the era dominated by dm and mcg type de .
for the critical point @xmath107 , the equation of state parameter given by equation ( 24 ) of the interacting de takes the form @xmath108^{-1-\alpha}\ ] ] + fig.1 + + + + + _ fig 1 : the dimensionless density parameters @xmath109 , @xmath110 and @xmath111 are plotted against each other in a 3d - scenario .
other parameters are fixed at @xmath112 and @xmath113 .
_ + fig.2 fig.3 fig.4 + + + + _ fig 2 : the dimensionless density parameters are plotted against e - folding time .
the initial conditions are @xmath114 and @xmath115 .
other parameters are fixed at @xmath116 and @xmath117 .
+ fig 3 : the dimensionless density parameters are plotted against e - folding time .
the initial conditions are @xmath118 and @xmath115 .
other parameters are fixed at @xmath119 and @xmath117 .
+ fig 4 : the dimensionless density parameters are plotted against e - folding time .
the initial conditions are @xmath118 and @xmath115 .
other parameters are fixed at @xmath120 and @xmath117 . _
+ fig.5 fig.6 fig.7 _ fig 5 : the phase diagram of the parameters @xmath121 and @xmath122 depicting an attractor solution .
the initial conditions chosen are @xmath123 ( green ) ; @xmath124 ( blue ) ; @xmath125 ( red ) ; @xmath126 ( gold ) .
other parameters are fixed at @xmath112 and @xmath117 .
+ fig 6 : the deceleration parameter is plotted against the eos parameter .
other parameters are fixed at @xmath112 and @xmath117 .
+ fig 7 : the statefinder parameter @xmath127 is plotted against the eos parameter .
other parameters are fixed at @xmath112 and @xmath117 . _ + + fig.8 fig.9 + _ fig 8 : the statefinder parameter @xmath128 is plotted against the eos parameter .
other parameters are fixed at @xmath112 and @xmath117 . +
fig 9 : the ratio of density parameters is shown against e - folding time .
the initial conditions chosen are v(1)=0.05 , u(1)=2.5 , y(1)=1.8 .
other parameters are fixed at @xmath112 and @xmath117 . _ + now we check the stability of the dynamical system ( eqs .
( 21 ) and ( 22 ) and ( 23 ) ) about the critical point . in order to do this
, we linearize the governing equations about the critical point i.e. , @xmath129 now if we assume @xmath130 , @xmath131 and @xmath132 then we may obtain @xmath133_{c}\delta u+\left[\partial_{v } f\right]_{c}\delta v+\left[\partial_{y } f\right]_{c}\delta y\ ] ] @xmath134_{c}\delta u+\left[\partial_{v } g\right]_{c}\delta v+\left[\partial_{y } g\right]_{c}\delta y\ ] ] and @xmath135_{c}\delta u+\left[\partial_{v } h\right]_{c}\delta v+\left[\partial_{y } h\right]_{c}\delta y\ ] ] where @xmath136 @xmath137 @xmath138 @xmath139 @xmath140 @xmath141 @xmath142 @xmath143 @xmath144 @xmath145 @xmath146 @xmath147 @xmath148 @xmath149 @xmath150 @xmath151 @xmath152 @xmath153 @xmath154 @xmath155 @xmath156 @xmath157 @xmath158 @xmath159 @xmath160 the jacobian matrix of the above system is given by , @xmath161 the eigen values of the above matrix are calculated at the critical point @xmath162 and are found to be + @xmath163 .
+ hence it is a * saddle point*. in ea model , the deceleration parameter @xmath164 can be obtained as @xmath165 @xmath166}\ ] ] from above we get , @xmath167 where @xmath168 @xmath169}\ ] ] if @xmath170 we have @xmath171 , which confirms the accelerated expansion of the universe .
when @xmath172 we have @xmath173 .
therefore we have super accelerated expansion of the universe .
+ in this scenario , the hubble parameter can be obtained as , @xmath174 where the integration constant has been ignored .
integration of eqn.(44 ) yields @xmath175 which gives the power law form of expansion of the universe . in order to have an accelerated expansion of universe in ea gravity we must have @xmath176 . using this range of @xmath177 in the equation @xmath178
, we get the range of @xmath179 as @xmath180 .
this is perfectly consistent with an accelerated expansion of the universe .
sahni et al ( 2003 ) @xcite introduced a pair of cosmological diagnostic pair @xmath181 which is known as as statefinder parameters .
the two parameters are dimensionless and are geometrical since they are derived from the cosmic scale factor alone .
also this pair generalizes the well - known geometrical parameters like the hubble parameter and the deceleration parameter .
the statefinder parameters are given by @xmath182 in ea model , we have the following expressions of @xmath127 and @xmath128 as @xmath183 and @xmath184
graphs are obtained and phase diagrams are drawn in order to determine the type of critical point obtained in this model .
below we discuss the results obtained in detail : the dimensionless density parameters @xmath109 and @xmath110 and @xmath111 are plotted against each other in figure 1 . in figs .
2,3 and 4 , the density parameters are plotted against time for different values of interacting coupling parameter . from the figures we see that @xmath109 always dominates over @xmath110 during evolution of the universe .
so this result is consistent with the well known idea of an energy dominated universe .
5 shows the phase portrait of the density parameters of de and dm .
as already stated before that the critical point obtained in this system is a saddle point and hence there always remains a question on the stability of the system . in fig.6 , a plot of deceleration parameter , @xmath164
is obtained against the eos parameter , @xmath185 .
it is seen that @xmath164 remains in the negative level thus confirming the recent cosmic acceleration
. figs . 7 and 8
show the plot of the statefinder parameters @xmath127 and @xmath128 respectively against the eos parameter , @xmath185 .
it is known that in case of @xmath1cdm model @xmath186 and @xmath187 . comparing figs . 7 and 8
, it can be seen that the values of @xmath127 is quite different from @xmath188 corresponding to the values of @xmath185 when @xmath187 .
this gives the deviation of the model from the @xmath1cdm model .
finally in fig.9 , the ratio @xmath189 is plotted against @xmath190 .
the decreasing trajectory confirms the existence of an energy dominated universe with progressive values of scale factor .
in this work , we have considered a combination of modified chaplygin gas in einstein - aether gravity model .
our basic idea was to study the background dynamics of mcg in detail when it is incorporated in aether gravity .
dynamical system analysis was carried out , critical points were found and the stability of the system around those critical points was tested . graphical analysis was done to get an explicit picture of the outcome of the work . in order to find a solution for the cosmic coincidence problem ,
a suitable interaction between de and dm was considered .
the set of equations characterizing the dynamical system was formed and a stable scaling solution was obtained .
hence this work can be considered to be a significant one as far as solution of cosmic coincidence problem is concerned . from the above analysis
we conclude that the combination of mcg in ea gravity makes a perfect model for the expanding universe undergoing a late acceleration .
the authors sincerely acknowledge the facilities provided by the inter - university centre for astronomy and astrophysics ( iucaa ) , pune , india where a part of the work was carried out .
the authors also thank the anonymous referee for his or her constructive comments that helped them to improve the quality of the manuscript .
+ 99 perlmutter , s. j. et al : - _ nature _ * 391 * 51 ( 1998 ) .
: - [ supernova search team collaboration ] _ astron .
j. _ * 116 * 1009 ( 1998 ) .
g. et al . : - _ astrophys . j. _ * 607 * 665 ( 2004 ) .
bennet , c. et al , : - _ phys .
lett . _ * 85 * 2236 ( 2000 ) .
spergel , d. n. et al .
: - _ astrophys
. j. suppl .
ser . _ * 170 * 377 ( 2007 ) .
adelman - mccarthy , j. k. et al .
: - _ astrophys
. j. suppl .
_ * 175 * 297 ( 2008 ) .
eisenstein , d. j. et al[sdss collaboration ] , 2005 , astrophys .
j. 633,560 .
briddle , s. et al : - _ science _ * 299 * 1532 ( 2003 ) .
spergel , d. n. et al .
: - _ astrophys
. j. suppl . _
* 148 * 175 ( 2003 ) .
peebles , p. j. e. , ratra , b. : - _ astrophys .
j. _ * 325 * l17 ( 1988 ) .
caldwell , r. r. , dave , r. , steinhardt , p. j. : - _ phys .
lett . _ * 80 * 1582 ( 1998 ) .
armendariz - picon , c. , mukhanov , v. f. , steinhardt , p. j. : - _ phys .
lett . _ * 85 * 4438 ( 2000 ) .
sen , a. : - _ jhep _ * 0207 * 065 ( 2002 ) .
caldwell , r. r. : - _ phys .
b _ * 545 * 23 ( 2002 ) . feng , b. , wang , x. l. , zhang , x. m. : - _ phys . lett .
b _ * 607 * 35 ( 2005 ) .
kamenshchik , a. y. , moschella , u. , pasquier , v. : - _ phys .
b _ * 511 * 265 ( 2001 ) .
debnath , u. , banerjee , a. and chakraborty , s. : - _ class .
quantum grav . _
* 21 * 5609 ( 2004 ) .
guo , z .- k . , zhang , y .- z .
: - _ phys . rev .
d _ * 71 * , 023501(2005 ) .
del campo , s. , herrera , r. , pavon , d. : - _ jcap _ * 0901 * 020(2009 ) .
alcaniz , j. s. , lima , j. a. s. : - _ phys .
d _ * 72 * 063516(2005 ) .
feng , c. et .
al . : - _ phys .
b _ * 665 * 111(2008 ) .
zlosnik , t.g . , ferreira , p.g . ,
starkman , g.d .
: - _ phys . rev .
d _ * 75 * 044017 ( 2007 ) zuntz , j. , zlosnik , t. g. , bourliot , f. , ferreira , p. g. , starkman , g. d. : - _ phys .
d _ * 81 * 104015 ( 2010 ) .
carroll , s. m. , lim , e. a. : lorentz violating vector field slow the universe down : - _ phys .
d _ * 70(12 ) * 123525 ( 2004 ) arxiv : hep - th/0407149 clifton , t. , ferreira , p. g. , padilla , a. , skordis , c. : - _ physics reports _ * 513 * ( 2012 ) chakraborty , s. , debnath , u. , ranjit , c. : - _ eur . phys .
c _ * 72 * 2101 ( 2012 ) ranjit , c. , rudra , p. , kundu , s. : - _ astrophys .
space sci . _ * 347 * 423 ( 2013 ) cohen , a. , kaplan , d. , nelson , a. : - _ phys .
lett . _ * 82 * 4971 ( 1999 ) .
sahni , v. , shtanov , y. : - _ jcap _ * 0311 * 014 ( 2003 ) .
cai , r. g. : - _ phys .
b _ * 657 * 228 ( 2007 ) .
wei , h. , cai , r. g. : - _ phys .
b _ * 660 * 113 ( 2008 ) .
choudhury , t. r. , padmanabhan , t. : - _ astron .
astrophys . _
* 429 * 807 ( 2007 ) .
tonry , j. l. et al .
: - _ apj _ * 594 * 1 ( 2003 ) .
barris , b. j. et al .
: - _ apj _ * 602 * 571 ( 2004 ) .
lu , j. et al .
: - _ phys .
b _ * 662 * 87 ( 2008 ) .
dao - jun , l. , xin - zhou , l. : - _ chin .
* 22 * 1600 ( 2005 ) .
dvali , g. r. , gabadadze , g. , porrati , m. : - _ phys .
b _ * 484 * 112 ( 2000 ) .
de felice , a. , tsujikawa , t. : - _ arxiv : 1002.4928 [ gr - qc]_. nojiri , s. , odintsov , s. d. : - _ arxiv:1011.0544 [ gr - q]_. clifton , t. , barrow , j. : - _ phys .
d _ * 72 * 103005 ( 2005 ) .
yerzhanov , k. k. et al . : - _ arxiv:1006.3879v1 [ gr - qc ] _ ( 2010 ) .
nojiri , s. , odintsov , s. d. : - _ phys .
b _ * 631 * 1 ( 2005 ) .
antoniadis , i. , rizos , j. , tamvakis , k. : - _ nucl . phys .
b _ * 415 * 497 ( 1994 ) .
horava , p. : - _ jhep _ * 0903 * 020 ( 2009 ) .
brans , c. , dicke , h. : - _ phys .
rev . _ * 124 * 925 ( 1961 ) .
rudra , p. , debnath , u. , biswas , r. : - _ astrophys .
space sci .
_ * 339 * 53 ( 2012 ) .
rudra , p. : - _ astrophys .
space sci .
_ * 342 * 579 ( 2012 ) .
chowdhury , r. , rudra , p. : - _ int .
phys . _ * 52 * 489 ( 2013 ) .
rudra , p. , ranjit , c. , kundu , s. : - _ astrophys .
space sci . _
* 347 * 433 ( 2013 ) .
zlosnik , t.g . , ferreira , p.g . , starkman , g.d .
: - _ phys . rev .
* 75 * 044017 ( 2007 ) .
zlosnik , t.g . , ferreira , p.g . , starkman , g.d .
: - _ phys . rev .
d _ * 77 * 084010 ( 2008 ) .
jacobson , t. , mattingly , d. : - _ phys .
d _ * 64 * 024028 ( 2001 ) .
jacobson , t. , mattingly , d. : - _ phys .
* 70 * 024003 ( 2004 ) .
garfinkle , d. , jacobson , t. : - _ phys .
lett . _ * 107 * 191102 ( 2011 ) .
linder , e. v. , scherrer , r. j. : - _ phys .
d _ * 80 * 023008 ( 2009 ) .
barrow , j. d. : - _ phys .
d _ * 85 * 047503 ( 2012 ) .
zuntz , j. , zlosnik , t.g . ,
bourliot , f. , ferreira , p.g . , starkman , g.d .
: - _ phys . rev .
d _ * 81 * 104015 ( 2010 ) .
li , b. , fonseca mota , d. , barrow , j.d .
: - _ phys . rev .
* 77 * 024032 ( 2008 ) .
gasperini , m. : - _ class .
* 4 * 485 ( 1987 ) .
gasperini , m. : - _ gen .
grav . _ * 30 * 1703 ( 1998 ) .
debnath , u. : - _ arxiv : 1310.2144v1 [ gr - qc ] _ meng , x. , du , x. : - _ phys .
b _ * 710 * 493 ( 2012 ) .
meng , x. , du , x. : - _ comm .
phys . _ * 57 * 227 ( 2012 ) .
sahni , v. et al .
: - _ jetp _ * 77 * 201(2003 ) . | in this work we investigate the background dynamics when dark energy is coupled to dark matter with a suitable interaction in the universe described by einstein - aether gravity .
dark energy in the form of modified chaplygin gas is considered . a suitable interaction between dark energy and dark matter
is considered in order to at least alleviate ( if not solve ) the cosmic coincidence problem .
the dynamical system of equations is solved numerically and a stable scaling solution is obtained .
a significant attempt towards the solution of the cosmic coincidence problem is taken .
the statefinder parameters are also calculated to classify the dark energy models .
graphs and phase diagrams are drawn to study the variations of these parameters .
it is also seen that the background dynamics of modified chaplygin gas in einstein - aether gravity is completely consistent with the notion of an accelerated expansion in the late universe .
finally , it has been shown that the universe follows the power law form of expansion around the critical point . 0.2 cm 0.2 cm 0.2 cm 0.2 cm 0.2 cm |
the experimental and theoretical investigations of reaction dynamics connected with the formation of composed system is nowadays the main subject of the nuclear reactions . at the first stage of reaction of heavy ion collisions the full momentum transfer can occur ( this event is defined as capture ) if there is a well in the nucleus - nucleus potential in dependence on the values of relative kinetic energy and friction coefficients @xcite . at capture , the two reacting nuclei form a rotating nuclear system at near coulomb barrier energies . during its evolution
this system can be transformed into compound nucleus or it re - separates into two fragments which may differ from the initial nuclei in the entrance channel @xcite . during the evolution of dns
its two nuclei may change their masses @xmath3 , @xmath4 and charges @xmath5 , @xmath6 but with constant total mass @xmath7 and charge @xmath8 . the dns should overcome the intrinsic fusion barrier @xmath9 ( it is equal to the difference between the maximum value of the driving potential and its value corresponding to the initial charge asymmetry ) to reach the compound nucleus state through more mass asymmetric configurations .
the intense of the break up of dns into two nuclei ( quasifission channel ) in competition with the complete fusion is characterized by the value of the quasifission barrier @xmath10 ( the depth of the pocket in the nucleus - nucleus potential ) @xcite .
the mass asymmetry parameter of quasifission fragments may be larger or smaller than that of nuclei in the entrance channel .
it is well known the case of the quasifission with small mass asymmetry in comparison with one of initial nuclei . in this case
the maximum of the mass distribution of quasifission fragments places between the masses of projectile ( or target ) and symmetric fission fragments .
on the other hand , due to the exchange with neutrons and protons between the nuclei constituting the dns the mass asymmetry parameter increases being larger than the mass asymmetry in the entrance channel ( evolution to complete fusion direction ) and at the same time there is also a relevant probability of dns to decay into two nuclei .
the decay probability depends on the excitation energy and quasifission barrier @xmath10 of dns at a given configuration .
of course , the mass distribution of quasifission products are related to the characteristics of the entrance channel ( beam energy and orientation angles of the axial symmetry axis of reacting nuclei ) .
the capture events surviving quasifission populate the complete fusion channel .
consequently , the deformed mononucleus may reach the fully equilibrated statistical shape of the compound nucleus ( cn ) , or if there is no the fission barrier @xmath11 the system immediately decays into two fragments ( fast fission process ) .
the latter phenomenon occurs only at high angular momentum @xmath12 for which the fission barrier of the complete fusion system disappears ( @xmath13= 0 ) .
therefore , the fast fission process takes place only in the reactions at high angular momentum values ( @xmath14 ) while the quasifission process takes place at all values of @xmath12 contributing to the capture reaction .
finally , in the last stage of nuclear reaction , the formed cn may de - excite by emission of light particles or undergoes to fission ( producing fusion - fission fragments ) .
the reaction products that survive fission are the evaporation residues ( er)@xcite . the registration of er is clear evidence of the compound nucleus formation , but generally the determination of er s only it is not enough to determine the complete fusion cross section and to understand the dynamics of the de - excitation cascade of cn if the fission fragments are not included into consideration . on the other hands ,
it is difficult for sure correct determination of the fusion - fission rate in the cases of overlapping of the mass and angle distributions of the products of different processes ( quasifission , fast fission and fusion - fission ) because sorting out the registered fissionlike fragments according to the mechanism of its origin is connected with some assumptions .
moreover , by observing the fission products of fissile nuclei formed in the in neutron ( or very light particles)-induced reactions with the heavy targets one can conclude that the low excited compound nucleus ( at about @xmath15 10 mev ) decays into very asymmetric fission fragments ( near to the shell closure ) , while the compound nuclei formed in heavy ion collisions at intermediate or high excitation energy ( @xmath16 20 mev ) undergo fission forming the mass symmetric fragments . starting from these general observations some researchers put forward the idea that the complete fusion process of two colliding nuclei may be considered as the inverse process to fission .
the authors of the papers @xcite argued that since the fission of a compound nucleus in heavy ion collisions produces just symmetric fragments , then in the collisions of two symmetric ( or almost symmetric ) nuclei complete fusion has to be a very probable process . but , unfortunately this is not true . for systems of colliding nuclei heavier than @xmath17pd+@xmath17pd
the fusion does not occur absolutely , while for the reactions like @xmath18mo+@xmath18mo , @xmath19zr+@xmath19zr , @xmath19zr+@xmath18mo , @xmath18mo+@xmath17pa or induced by projectiles higher than zn , ge , kr there is a strong hindrance to fusion .
following the previous reasons one can affirm that the hypothetical @xmath20sn+@xmath21cd reaction should lead to the @xmath22cf cn since @xmath21cd ( with @xmath23=48 near the shell closure 50 ) and @xmath20sn ( with double shell closure @xmath23=50 and @xmath24=82 ) are produced with highest yields in spontaneous fission of @xmath22cf .
but our estimation for this reaction does not give meaningful fusion probability ( @xmath25 ) .
the simple reason resides in the peculiarities of the reaction dynamics . in the spontaneous fission of @xmath22cf the average value of angular momentum distribution of the fragments
is close to zero , but if we want to reach the @xmath22cf compound nucleus , by the hypothetical @xmath20sn+@xmath21cd reaction ( or by the realistic @xmath20sn+@xmath26cd reaction leading to the @xmath27cf cn ) , the average value of angular momentum distribution of dns in the entrance channel may be about @xmath28 or higher by increasing the beam energy .
the @xmath29 value is calculated as @xmath30 in the fusion reaction mechanism , at the first stage , dns which is formed after capture of projectile by the target nuclei should survive quasifission ( re - separation of nuclei of dns ) . due to the hindrance connected with quasifission for these system
the fusion probability @xmath31 should be lower than 10@xmath32 for small values of angular momentum .
this low probability of complete fusion becomes more lower for the excited and fast rotating deformed mononucleus which undergoes fast fission before the system can reach the compact shape of compound nucleus . also in the cases of the explored @xmath33ne+@xmath34cf ( more mass asymmetric system ) , @xmath35mg+@xmath27 cm , @xmath36si+@xmath37pu , @xmath38s+@xmath39u , and @xmath40ar+@xmath41th ( less mass asymmetric system ) reactions ,
the compound nucleus @xmath42hs is formed with different angular momentum distributions in dependence on entrance channels even the defined excitation energy @xmath43 has been achieved .
therefore , such compound nuclei decay by different yields of reaction products .
the fusion - fission fragment mass distributions are peaked at around the @xmath44xe nucleus with different dispersions and average angular momentum distributions in connection with the various entrance channels .
if we calculate the formation probability of the @xmath42hs compound nucleus in the mass symmetric @xmath44xe+@xmath44xe reaction at the same fixed excitation energy @xmath45 as in the considered @xmath33ne+@xmath34cf reaction ( where @xmath46 1 ) , we do not meaningfully reach such a compound nucleus(@xmath4710@xmath48 ) .
the angular momentum distribution for the @xmath44xe+@xmath44xe collision at the capture stage is completely different and all conditions of reaction dynamics lead to deep inelastic and quasifission products . in this context , for the @xmath44xe+@xmath20sn ( @xmath4710@xmath49 ) and @xmath20sn+@xmath50yb ( @xmath475@xmath5110@xmath52 ) reactions , one can observe the same above - described hindrance to complete fusion .
our theoretical capture cross section @xmath53 includes all damped reactions , excluding deep inelastic collisions ( dic ) .
the partial capture cross sections are contributed by the full momentum transfer events corresponding to the trapping into a pocket in the nucleus - nucleus potential after dissipation of the relative kinetic energy .
differently , the dic events are not characterized by the full momentum transfer and collision paths are not trapped in the pocket of the nucleus - nucleus potential .
the dic events are not connected with any pocket of a potential ; for example at large values of @xmath12 there is no pocket but the dic events take place .
the partial capture cross section is calculated by the following formula @xmath54 with the capture probability @xmath55 which is found by the solution of the motion equations @xcite .
[ prima ] in the calculation of nucleus - nucleus potential , which includes the coulombian @xmath56(@xmath23 , @xmath57 , @xmath58 ) , nuclear @xmath59(@xmath23 , @xmath57 , @xmath58 ) and rotational @xmath60(@xmath23 , @xmath57 , @xmath58 , @xmath12 ) parts , we take into account the static and dynamic deformations , and orientation angles of the axial symmetry axes of reacting nuclei , at initial stage . in fig .
[ seconda ] we present an example of capture events with different potential wells and its dependencies on the angular momentum and orientation angles of axes of reacting nuclei ; moreover , in figures we also present the cases of capture and dic events by changing the beam energies and angular momentum values .
the lifetime of an excited dns for a given reaction depends on the initial collision energy @xmath61 and angular momentum distribution values ( see formulae ( 6)-(8 ) and fig.4 of our paper ) .
the mass and charge , as well as shapes of nuclei constituting the dns are changed during interaction time .
the evolution of dns depends on its excitation energy , orientation angles of the axial symmetry axes and shell structures of reacting nuclei . during its evolution dns
can evolve to complete fusion or can decay into two fragments ( quasifission process ) .
the competition between these two processes is related to the values of intrinsic fusion barrier @xmath9 and quasifission barrier @xmath10@xcite depending on the peculiarities of reacting nuclei , beam energy and angular momentum distribution . in many cases and conditions
the mass and angular momentum distributions of quasifission fragments can overlap with the mass and angular distributions of fusion - fission fragments , or / and with the ones of the fast fission fragments@xcite , leading to the real difficulties in the experimental analysis in order to sort out the true yields of fragments belonged to various reaction mechanisms ( see for example @xcite and fig .
@xcite ) .
[ figjpcs3 ] usually , in reactions with actinides the maximum of the yield of the quasifission products is observed in the mass region between masses of the projectile - like products and symmetric masses . in the @xmath0ca + @xmath62pb reaction ,
the maximum of the quasifission products is very close to the region of projectilelike and targetlike fragments ( see fig . 3 of ref .
@xcite , as well as fig . 1 of ref . @xcite and fig
. 8 of ref .
this is due to the fact that projectile and target are double magic nuclei . as a result
the mass distributions of the capture , _
i.e. _ quasifission reactions , and deep inelastic collisions are mixed .
the reaction time of the deep inelastic collisions is much smaller than that of quasifission and fast fission reactions , because the last two reactions take place if capture occurs , _
i.e. _ projectile has been trapped into potential well ( see fig .
[ prima ] ) .
it is difficult to separate without uncertainty products of these two processes in experimental data . to demonstrate the importance of this circumstance
, the mass distributions of the deep - inelastic and capture reactions were calculated by the method developed on the basis of the dns concept @xcite .
the results for the @xmath0ca+@xmath27pb reaction are presented in fig .
[ fig4mpla ] .
the reason of this result can be seen from the driving potential which has a deep valley around initial charge asymmetry due to the nuclear shell effects .
therefore , the sufficient part of the yields of quasifission and deep - inelastic products are mixed in the mass and charge distributions .
the difference between them is in their angular distributions which are connected with the lifetime of the formed rotated dns .
the overlap of the maximum of the mass distribution of quasifission with that of the deep inelastic collisions can lead to conclusion that there is no the contribution of the quasifission process in the total reaction cross section .
this explains the strong deviation of the theoretical values of the capture cross section at low energies from the experimental capture cross section which was determined by the symmetric fragmentation only .
so , the mixing of the mass distributions of the deep inelastic collisions and quasifission processes causes an ambiguous determination of the capture cross section from the experimental mass distributions in case of reactions with magic or double magic nuclei , like to the case of the reaction under discussion . as well as the mixing of the mass distributions of the quasifission , fast fission and fusion - fission reactions makes difficult the estimation of the complete fusion cross section leading to the genuine compound nucleus .
due to the strong dependence of the mass distribution on the peculiarities of nuclear shell structure , the proportion of mixing is changed from one reaction to another reaction .
the theoretical study of this problem is useful to establish the reactions and beam energies which are favorable to obtain the maximum cross section of formation of the compound nucleus with the relatively low angular momentum and excitation energy .
we can state that , at the low energies , the products of the symmetric fragmentation in the @xmath0ca+@xmath62pb reaction are generally formed at usual fission of cn , and therefore they are related to the fusion cross section . at energies
@xmath63 mev the contribution of the fast fission appears and it increases by the beam energy .
that will be shown in fig .
[ fasfis ] .
moreover , fig .
[ figjpcs2 ] shows the mass distribution of the quasifission products of the @xmath0ca+@xmath64sm reaction as a function of the dns lifetime , following the capture reaction at two @xmath61 energies of 140 mev ( panel ( a ) ) and 160 mev ( panel ( b ) ) . at lower energies ( panel ( a ) ) the mass distribution of quasifission fragments are asymmetric ; at higher beam energy ( panel ( b ) ) , also appear mass symmetric quasifission fragments .
the latter mass distribution of quasifission fragments overlap to the fusion - fission mass distribution leading to a relevant uncertainty in experimental determination of the fusion cross section . in fig .
[ figjpcs3 ] we present the results of the @xmath0ca+@xmath64sm reaction regarding the determination of the rotational angle @xmath65 of the dinuclear system at two @xmath61 beam energies of 138 and 154 mev ( see panels ( a ) and ( b ) , respectively ) versus the angular momentum @xmath66 values of dns , and the angular distributions of the quasifission @xmath67cr and @xmath68ce fragments at decay of dns for the energies @xmath61=138 and 154 mev , ( see panels ( c ) and ( d ) , respectively ) .
as one can see , at @xmath61=138 mev the decay of dns into the two mentioned fragments lead to large rotational angle @xmath65 at intermediate and high angular momentum components ( panel ( a ) ) as well as a wide angular distributions of the quasifission @xmath67cr and @xmath68ce fragments at all dns rotational angles ( panel ( c ) ) ; while , at high energy @xmath61=154 mev , the rotational angle @xmath65 is small up to large angular momentum values ( panel ( b ) ) , and the angular distribution of the quasifission fragments is strongly peaked to low rotational angle values of dns ( panel(d ) ) .
this means that at @xmath61=138 mev for @xmath69 35@xmath70 the rotation angle is enough large and the angular distribution of the quasifission fragments may be nearly isotropic .
the maximum of the angular distribution moves to small angles at larger energies : @xmath61=154 mev the most part of fragments is distributed to the forward angles and to backward angles for the relative fragment - partner ( see panel ( d ) ) , and the maximum of the angular distribution are concentrated around @xmath65 = 15@xmath71 in the center - of - mass system .
therefore the large part of quasifission fragments passes beside the nearest to the beam detector .
this phenomenon caused the conclusions of the authors of ref .
@xcite about decreasing of quasifission events at energies @xmath72154 mev .
the reason of decreasing in the experimental events of quasifission at low energies @xmath73138 mev is the isotropic angular distribution of some part of quasifission fragments which were considered as the compound nucleus fission fragments .
the composite system formed at capture stage evolves by exchanging nucleons between the two nuclei constituting the dns . during its evolution
the dns can reach the shape of a deformed mononucleus ( complete fusion ) or it can break up into two fragments ( quasifission ) without reaching the complete fusion stage . in the first case
the nuclear system has to reach the statistical equilibrate shape of the cn with the angular momentum @xmath74 .
the equilibrium shape of compound nucleus depends on the value of @xmath75 which determines the fission barrier @xmath11 providing its stability against to fission .
it is well known that @xmath11=0 if the value of @xmath75 is larger than its critical value @xmath76 for the compound nucleus with the given mass and charge number @xcite .
therefore , the events of complete fusion for @xmath76 can not reach the equilibrium shape of cn because the mononucleus immediately decays into two fragments . this fissionlike process is called fast fission because compound nucleus is not formed .
competition between complete fusion and quasifission occurs at all values of the orbital angular momentum .
therefore , the partial capture cross section is contributed by the following terms : @xmath77 where @xmath78 is the partial fusion cross section and @xmath79 is fusion probability as a function of @xmath12 .
it is clear that the fusion cross section in formula ( [ s_fus ] ) includes the cross sections of evaporation residue and fusion - fission products .
the fusion cross section , for each orientation angles of the symmetry axes of the deformed reacting nuclei , is obtained by formula @xmath80 therefore , taking into account the contributions of all configurations with the orientation angles of the symmetry axes of the deformed reacting nuclei we can calculate the averaged value of the fusion cross section by formula : @xmath81 when both nuclei are deformed nuclei , or by formula : @xmath82 for the spherical projectile and deformed target .
obviously , the quasifission cross section is obtained as @xmath83 where @xmath84 is the maximum value of @xmath12 at which the capture events occur .
another binary process which leads to the formation of two fragments similar to the ones of fusion - fission or quasifission is the fast fission .
the fast fission cross section is calculated by summing the contributions of the partial cross sections related to the range @xmath85 ( for which @xmath11=0 ) leading to the formation of the mononucleus : @xmath86 the fission and the evaporation residue cross sections are calculated by the advanced statistical code @xcite that takes into account the damping of the shell correction in the fission barrier as a function of nuclear temperature and orbital angular momentum . the fusion cross section is calculated from the branching ratio @xmath87 of the decay rates of overflowing the border of the potential well ( @xmath88 ) along @xmath58 at a given mass asymmetry ( decay of dns in quasifission fragments ) over the barriers on mass asymmetry axis @xmath89 for the complete fusion or @xmath90 in opposite direction to the symmetric configuration of dns ( for details see fig . 4 of ref .
@xmath91 where @xmath92 , @xmath93 and @xmath94 are corresponding widths determined by the level densities on the barriers @xmath89 , @xmath95 and @xmath10 involved in the calculation of @xmath31 are used in the model @xcite based on the dinuclear system concept @xcite . here
@xmath96 is the excitation energy of dinuclear system in the entrance channel , where @xmath97 and @xmath98 are charge and mass numbers of the projectile nucleus .
@xmath99 is the minimum value of the nucleus - nucleus potential well ( for the dns with charge asymmetry @xmath23 ) and its position on the relative distance between the centers of nuclei is marked as @xmath100 .
the value of @xmath10 for the decay of dns with the given charge asymmetry of fragments is equal to the depth of the potential well in the nuclear - nuclear interaction .
the intrinsic fusion barrier @xmath89 is connected with mass ( charge ) asymmetry degree of freedom of the dinuclear system and it is determined from the potential energy surface : @xmath101 here , @xmath102 , @xmath103 and @xmath104 are the binding energies of the nuclei in dns and the cn , respectively , which were obtained from @xcite ; the fragment deformation parameters @xmath105 are taken from the tables in @xcite and @xmath106 are the orientation angles of the reacting nuclei relative to the beam direction ; @xmath107 is the rotational energy of the cn .
the distribution of neutrons between two fragments for the given proton numbers @xmath23 and @xmath108 or ratios @xmath109 and @xmath110 for both fragments were determined by minimizing the potential @xmath111 as a function of @xmath57 for each @xmath23 .
the driving potential @xmath112 is a curve linking minima in the potential well corresponding to each charge asymmetry @xmath23 in the valley of the potential energy surface from @xmath113 up to @xmath114 as a function of the relative distance ( see fig . 4 of ref .
we define the intrinsic fusion barrier for the dinuclear system with charge asymmetry @xmath23 as @xmath115 , where @xmath116 is a maximum value of potential energy at @xmath117 in the valley along the way of complete fusion from the given @xmath23 configuration .
the @xmath118 is defined by the similar way as shown in fig .
4 c of ref .
@xcite .
the masses and charges of the projectile and target nuclei are dynamical variables during capture and after formation of the dns .
the intense proton and neutron exchange between constituents of dns is taken into account by calculation of the complete fusion probability @xmath31 as fusion from all populated dns configurations according to the formula @xmath119 where @xmath120 is the excitation energy of dns with angular momentum @xmath12 for a given value of its charge - asymmetry configuration @xmath23 and @xmath121 ; @xmath122 ; @xmath123 is the change of @xmath124-value by changing the charge ( mass ) asymmetry of dns ; @xmath125 is the probability of population of the ( @xmath126-z ) configuration at @xmath127 and given orientation angles ( @xmath128 ) .
@xmath129 is the probability of population of the configuration @xmath130 at @xmath131 and @xmath12 .
the evolution of @xmath132 is calculated by solving the transport master equation : @xmath133 here , the transition coefficients of multinucleon transfer are calculated as in @xcite @xmath134 where the matrix elements @xmath135 describe one - nucleon exchange between the nuclei of dns , and their values are calculated microscopically using the expression obtained in ref . @xcite .
a non - equilibrium distribution of the excitation energy between the fragments was taken into account in calculations of the single - particle occupation numbers @xmath136 and @xmath137 as it was done in ref .
@xcite ; @xmath138 and @xmath139 are perturbed energies of single - particle states . in eq .
[ massdec ] , @xmath140 is the kramers rate for the decay probability of the dinuclear system into two fragments with charge numbers @xmath23 and @xmath141 ( details in ref .
@xcite ) , and it is proportional to @xmath142 where @xmath143 is the quasifission barrier .
[ pcnecm ] eqs .
( [ massdec ] ) with the coefficients ( [ delt ] ) and initial condition @xmath144 are solved numerically and the primary mass and charge distributions are found for a given interaction time @xmath145s ( see ref .
@xcite ) . in eq .
( [ pcny ] ) , we use the definition @xmath146 . in ( [ delt ] )
we use @xmath147s@xmath148 . as an example of our calculation of the fusion probability @xmath31 versus the angular momentum @xmath12 , we present the results for the @xmath0ca+@xmath64sm reaction at two different @xmath45 values ( see fig .
[ pcnfig ] ( a ) ) and the function @xmath31 versus the energy @xmath149 for the reaction that lead to different isotopes of thorium ( see fig .
[ pcnfig ] ( b ) ) .
this last figure shows the effect of the entrance channel on the @xmath31 fusion probability
. moreover , fig .
[ pcnecm ] shows the calculation of @xmath31 as a function of the energy @xmath61 for the two close @xmath150s+@xmath151w and @xmath150s+@xmath152w reactions for a wide range of excitation energies , in comparison with the calculation reported in ref .
for these two close reactions we observe two region of strong hindrance to fusion at very small and large values of the collision energy of @xmath61 for both reactions while the approximate formula used in ref .
@xcite gives similar results only at small energies , but @xmath31 increase quickly reaching a maximum value equal to 1 at @xmath61 = 145 mev .
in addition , there is no difference between the values of @xmath31 calculated by the approximated formula for the @xmath150s+@xmath151w and @xmath150s+@xmath152w reactions .
the branching ratio of the quasifission , fusion , and fast fission cross sections in formula ( [ capt2 ] ) depend on the masses and charges , shell structure , mass asymmetry of projectile - target nuclei , as well as on the energy and impact parameter of heavy ion collision . as it was noted in section [ intro ] , the formation of quasifission and fast fission products bypasses the stage of the cn formation .
the difference between the two mentioned processes is in their dependence on the orbital angular momentum : quasifission is possible at all angular momenta of collision as fast fission occurs only at @xmath153 , where @xmath154 is the angular momentum value at which the fission barrier of the compound nucleus disappears .
[ fig1 ] the mass distributions of products of the capture reactions can be mixed by different proportions in different mass regions . to avoid ambiguity , the authors of ref .
@xcite referred to the registered products as ones of the symmetric fragmentation ( capture ) because the genuine fission process is the decay of the compound nucleus into two fragments .
the mass distribution of products of the fast fission reactions can be alike to that of the fusion - fission reactions .
but the angular distribution of products of the former reaction is more anisotropic than that of the latter reaction due to the difference in their reaction time . in order to estimate the contribution of the fast fission into the calculated capture cross section
, we use the value @xmath155 obtained in ref .
@xcite for the @xmath0ca+@xmath62pb reaction . in fig .
[ fasfis ] the results of calculation for the fusion angular momentum distribution of this reaction is presented .
one can see that maximum of the partial fusion cross section is around @xmath156 .
the fast fission contribution for @xmath157 is calculated by formula ( [ udz ] ) , while the fusion cross section of cn must include the evaporation residues and fusion - fission cross sections only . in our calculation of evaporation residues , the values of the angular momentum @xmath158
were used .
the probability of the formation of a evaporation residue nucleus surviving fission with mass number @xmath159 and charge number @xmath160 from a heated and rotated compound nucleus @xmath161 after emissions of @xmath162 neutrons , @xmath163 protons , and @xmath164-particles at the @xmath165th step of the de - excitation cascade is determined by the formula @xcite : @xmath166 where @xmath167 is the partial cross section of the intermediate nucleus formation at the @xmath168th step and @xmath169 is the survival probability of the @xmath168th intermediate nucleus against fission along the de - excitation cascade of cn ; @xmath154 is the value of angular momentum @xmath12 at which the fission barrier for a compound nucleus disappears completely @xcite ; @xmath170 is the excitation energy of the nucleus formed at the @xmath165th step of the de - excitation cascade .
it is clear that @xmath171 at @xmath172 where @xmath61 , @xmath173 , and @xmath174 are the collision energy in the center of mass system , the reaction @xmath173-value , and rotational energy of the compound nucleus , respectively .
the numbers of the emitted neutrons , protons , @xmath175-particles and @xmath176-quanta , @xmath177n , @xmath178p , @xmath179 , and @xmath180 , respectively , are functions of the step @xmath165
. the emission branching ratios of these particles depend on the excitation energy @xmath181 and angular momentum @xmath182 of the cooling intermediate nucleus .
[ pcnorien ] the de - excitation cascade , characterized by the emission of the above - mentioned particles , starts from the compound nucleus @xmath161 .
its formation probability is the partial cross section of complete fusion @xmath183 corresponding to the orbital angular momentum @xmath12 .
the fusion cross section is equal to the capture cross section for light systems or light projectile - induced reactions , while for reactions with massive nuclei , it becomes a model - dependent quantity .
concerning estimation of the fusion cross section from the experimental fragments data , its value is sometimes an ambiguous quantity because of difficulties in separating the fusion - fission fragments from the quasifission fragments in the case of overlap of mass and angular distributions . as an example we present the results of our calculation for the @xmath0ca+@xmath64sm reaction regarding the fusion probability @xmath184 ( see fig .
[ pcnorien ] ( a ) ) and evaporation residue cross sections @xmath185 ( see fig .
[ pcnorien ] ( b ) ) as a function of the collision energy @xmath186 for different values of the target orientation angles @xmath187 .
the dependence of the quasifission - fusion competition during the evolution of the dinuclear system and the sensitivity of the fission - evaporation competition during the de - excitation cascade of the compound nucleus on the values of orientation angle @xmath188 are demonstrated .
the analysis of the dependence of the compound nucleus and evaporation residue formation cross sections on @xmath188 shows that the observed yield of evaporation residues in the @xmath0ca+@xmath64sm reaction at the low energies ( @xmath73137 mev ) is formed in the collisions @xmath189 . because the initial beam energy is enough to overcome the corresponding coulomb barrier for the collisions with these orientation angles @xmath188 . only in this case
it is possible formation of dinuclear system which evolves to compound nucleus or breaks up into two fragments after multinucleon exchange without formation of the compound nucleus . at larger energies ( about @xmath190140180 mev )
all orientation angles of the target - nucleus can contribute to @xmath191 and its values are in the 10100 mb range .
at the larger collision energies ( @xmath192158 mev ) the complete fusion still increases but the evaporation residue cross section @xmath191 goes down and its values are in the 10.1 mb range due to the strong decrease of the survival probability of the heated compound nucleus along de - excitation cascade .
this is connected by the decrease of the fission barrier for a compound nucleus by increasing its excitation energy @xcite and angular momentum @xcite .
another phenomenon leading to decrease of @xmath191 at higher energy is the fast fission process which is the splitting of the mononucleus into two fragments due to absence of the fission barrier at very high the angular momentum @xmath193 .
to show the importance of the overlap of mass distribution of quasifission and fusion - fission products in the unambiguous sorting out of events belonging to true fusion - fission process from the ones of quasifission process , we compare theoretical and experimental results of capture and fusion cross section for the @xmath0ca+@xmath27 cm reaction in fig .
[ 116cf ] and discuss these results .
the calculated capture cross section ( solid line in fig.[116cf ] ) is in good agreement with the experimental data @xcite for the production of all fragments ( full squares ) , while the theoretical fusion cross section ( dashed line ) is not in agreement with the experimental data obtained assuming the masses of the fusion - fission fragments belong in the ranges @xmath194 or ( @xmath195 .
the presence of the quasifission contribution into measured fusion - fission cross section in the ranges under discussion causes disagreement with the theoretical results . if it is assumed that the experimental fusion - fission events are in the @xmath194 amu interval ( almost close to the @xmath196 value ) , the calculated fusion cross sections ( the dashed line ) will be closer to the new set of the experimental data ( open triangles @xcite in fig.[116cf ] )
indeed , in this case there is an appreciable contribution of the quasifission process ( or a contribution which can not be neglected ) , in addition to the fusion - fission fragment formation .
therefore , the estimated experimental fusion cross section , connected with the new set of the experimental events of fission fragments , still appears to be a little larger than the calculated fusion excitation function at higher excitation energies .
a preliminary calculation of the mass distribution of quasifission fragments for a fixed reaction time @xmath197 of a dns performed in the framework of the model developed on the basis of the dinuclear system concept @xcite , indicates that the fragments of the quasifission process also appear in the mass - symmetric region and are mixed with the fragments coming from the fusion - fission process .
there are three main processes causing hindrances to er formation in reactions with massive nuclei : quasifission , fusion - fission , and fast fission @xcite .
all of these processes produce binary fragments in different stages of reaction .
moreover , the angular and mass distributions of some parts of their products can overlap @xcite .
ignoring this mixing may lead to ambiguity at analysis of the experimental data connected with the binary fragments .
this problem should be studied carefully .
the er formation process is often considered as the third stage of the three - stage process . the first stage is capture formation of the dns after full momentum transfer into the deformation energy of nuclei , their excitation energy , and rotational energy from the initial relative motion of the colliding heavy ions in the center - of - mass system .
capture takes place if the initial energy of the projectile in the center - of - mass system is enough to overcome the interaction barrier ( coulomb barrier + rotational energy of the entrance channel ) @xcite .
the study of the dynamics of heavy ion collisions at energies near the coulomb barrier shows that complete fusion does not occur immediately in collisions of massive nuclei @xcite .
after formation of the dns , the quasifission process competes with the formation of cn .
quasifission occurs when the dns prefers to break down into fragments instead of being transformed into a fully equilibrated cn .
the number of events contributing to quasifission increases drastically by increasing the sum of the coulomb interaction and rotational energy in the entrance channel @xcite .
another reason for the decreasing yield of er with increasing excitation energies is the usual fission of a heated and rotating cn that was formed in competition with quasifission . the stability of a massive cn decreases due to the decrease in the fission barrier by increasing its excitation energy @xmath198 and angular momentum @xmath12 @xcite .
comparison of the experimental values of the evaporation residue cross sections normalized with respect to the capture cross sections for the @xmath199o+@xmath152w ( solid circles ) @xcite and @xmath200f+@xmath201ta systems ( solid squares ) @xcite with the corresponding theoretical results ( dashed and solid lines , respectively ) as a function of the excitation energy @xmath45 of cn ( left axis ) .
theoretical results of the sum of the quasifission and fast fission cross sections ( normalized with respect of the fusion cross sections ) for the @xmath199o+@xmath152w ( dot dashed line ) and @xmath200f+@xmath201ta ( dot - dot dashed line ) systems are presented versus @xmath45 and compared on the right axis.,scaledwidth=80.0% ] the theoretical values of the quasifission partial cross sections for the @xmath200f+@xmath201ta and @xmath199o+@xmath152w reactions are presented in the left panels of fig .
it is seen from these figures that the quasifission takes place at all values of @xmath12 leading to capture .
the angular momentum distributions of cn formed in these reactions at the excitation energies @xmath20272 and 91 mev are presented in the right panels of fig .
the spin distributions of cn formed in each of these reactions differ mainly by the probability but not by the values of the angular momentum ranges .
this means that the number of cn formed in both reactions under discussion are different , but they have a similar range of the angular momentum @xmath12 .
the vertical dotted lines at @xmath203 in these panels separates the complete fusion ( @xmath204 ) and fast fission ( @xmath205 ) regions of the angular momentum .
the quasifission and fast fission processes produce binary fragments which can overlap with those of the fusion - fission channel and the amount of mixed detected fragments depends on the mass asymmetry of entrance channel , as well as on the shell structure of the reaction fragments being formed .
the suggestion for the experimental studies of the difference between characteristics of the fusion - fission , quasifission and fast fission products can be made when their mass ( charge ) , kinetic energy and angular distributions are explored in detail by dynamical calculations allowing to obtain the relaxation times of these processes .
therefore , the correct estimation of the cn formation probability in the reactions with massive nuclei is a difficult task for both experimentalists and theorists .
different assumptions about the fusion process are used in different theoretical models which can give different cross sections .
the experimental methods used to estimate the fusion probability depend on an unambiguous identification of the complete fusion products among the quasifission products .
the difficulties arise when the mass ( charge ) and angular distributions of the quasifission and fusion - fission fragments strongly overlap , depending on the reaction dynamics . as a result
, the complete fusion cross sections may be overestimated @xcite .
we confirm that the compared ratios of the cross sections between evaporation residues and complete fusion @xmath206 for the @xmath199o+@xmath152w and @xmath200f+@xmath201ta reactions discussed in @xcite are not free from the above - mentioned ambiguity in the determination of the fusion cross section @xmath207 .
theoretical values of the fusion cross section include only evaporation residues and fusion - fission cross sections @xmath208 the experimental values of fusion cross section reconstructed from the detected fissionlike fragments and evaporation residues @xcite : @xmath209 where @xmath210 , @xmath211 , and @xmath212 are the contributions of fusion - fission , quasifission and fast fission processes , respectively , and @xmath213 is the er contribution . according to the statement of the authors of ref .
@xcite , the complete fusion cross sections are obtained by adding fission cross sections @xcite to the measured data of the evaporation residue cross sections@xcite .
therefore , we can state that the definition of the experimental fusion cross section is similar with the definition of capture @xcite : @xmath214 . in ref .
@xcite , the complete fusion cross section is derived from a statistical model where only neutron evaporation and fission are included .
we think that the fission data from ref .
@xcite contain quasifission fragments and , at larger beam energies , also fast fission contributions , which appear as hindrances to complete fusion .
this argument is confirmed by our results obtained in the framework of the dns model .
we calculate the total er and fusion - fission excitation functions in the framework of the advanced statistical model @xcite .
by using the above - mentioned method we calculated the capture , quasifission , fast fission and fusion ( formation of cn ) cross sections for the @xmath0ca induced reaction on the @xmath1bk target ( see fig .
[ production ] ( b ) ) . in the @xmath73200 mev energy range
the fusion cross section is about three orders of magnitude lower than that of quasifission ( which is close to the capture cross section in the whole explored energy range ) , and it is about one order of magnitude lower than the one of fast fission . at higher @xmath61 energies , the fusion cross section is about two orders of magnitude lower than that of quasifission , while it is about five times lower than the one of fast fission . in this reaction
the quasifission is the completely dominant process in comparison to complete fusion , and the fast fission is also a strongly relevant process in comparison to the fusion formation ( cn ) . in this case ,
fast fission process takes place starting from the @xmath216 value equal to 30 @xmath70 . as a function of the excitation energy for the @xmath0ca+@xmath1bk reaction leading to @xmath215117 compound nucleus.,scaledwidth=83.0% ]
the shell corrections for the odd - odd and odd - even nuclei in the 292 - 297 mass region of the 117 element were obtained from the static fission barrier @xmath11 ( for @xmath12=0 ) calculated by sobiczewski@xcite , which correspond to the realistic values of shell corrections , as an extension of the results presented in the paper of kowal _ et al . _
@xcite for the even - even superheavy nuclei .
we calculated the average effective fission barriers and evaporation residue cross sections for all nuclei along the de - excitation cascade of the @xmath215117 compound nucleus by using such values of shell corrections ( for @xmath12=0 ) and taking into account their fade - out ( see refs . @xcite and references therein ) as a function of the temperature and angular momentum . in fig .
[ production](b ) , we compare the individual excitation functions of the evaporation residue for emission of neutron only along the cascade of cn with er cross sections after 3n and 4n emission from cn measured by oganessian _
_ in their experiment@xcite
. as one can see our result are in good agreement with the experimental data and we also find that the peaks of such excitation functions for 3n and 4n emission are present at about 30.6 and 34.0 mev , respectively .
the maximum yield of the 2n contribution ( @xmath217pb ) is lower than the ones of the 3n and 4n contributions .
the maximum yield of the 5n contribution is about 2.3 @xmath218 pb at @xmath198= 43.2 mev .
moreover , in fig . [ figbarr ]
we report the effective fission barrier @xmath2 distribution as a function of @xmath219 along the cascade of cn at different excitation energies and neutron number of the intermediate excited nuclei . in table
[ tab1 ] are reported such @xmath2 values obtained for the chain of nuclei , at three different @xmath220 values of the considered excited nuclei .
.effective fission barrier @xmath2 for the @xmath0ca+@xmath1bk@xmath221117 reaction leading to intermediate excited nuclei with masses @xmath57=297 , 296 , 295 , 294 , 293 and 292 , at excitation energies of nuclei @xmath219= 17 , 25 and 33 mev . [ cols="^,^,^,^,^,^,^",options="header " , ] so , all above - mentioned reactions are excellent reactions for the study of the quasifission process . for the last reaction ( @xmath39u+@xmath27 cm ) in table [ tab2 ] the @xmath31 value is not reported because for this reaction capture cross section does not exist .
moreover , at energies higher than the coulomb barrier , the fast fission process gives a relevant contribution to formation of fissionlike fragments .
for these reactions the contribution of the complete fusion cross section to the capture cross section is very small .
consequently , the er cross section is very small too , or in many cases is lower than the limit ( about 0.2 pb ) of the present possibility of the experimental setup .
therefore , it is necessary to estimate the quasifission , fast fission and complete fusion cross sections by a hopeful analysis of the reaction dynamics before to plan new long beam - time and expensive experiments . because only the fusion - fission and fusion - evaporation yields contribute to the complete fusion cross section . only in the two above - mentioned reactions ( @xmath222zr+@xmath20sn , @xmath223kr+@xmath44xe ) it is possible to observe evaporation residues . for all other reactions listed in table [ tab2 ]
it is not possible to observe meaningful fusion - fission contributions and least of all evaporation residues .
the ratio of fusion - fission and quasifission fragment yields for the all other mentioned reactions are much lower than 10@xmath224 .
therefore it is impossible to separate the fragments of fusion - fission process from the fragments of the huge amount of quasifission process .
our theoretical model is a powerful predictive method to calculate the yield of reaction products and also to describe the processes in the reactions with massive nuclei .
moreover , also for reactions used to synthesis of superheavy elements our model is able to calculate the effective fission barriers along the various steps of the de - excitation cascade with the neutron and proton emission from the compound nucleus and intermediate excited nuclear systems at various excitation energy and angular momentum .
60 volkov v v in contributed papers of nucleus - nucleus collision ii , visby , 1985 , edited by b. jakobson and k. aleclett ( north - holland , amsterdam , 1985 ) , vol.1 , p.54 ; izv .
nauk sssr ser .
50 , 1879 ( 1986 ) ; in proceedings of international school - seminar on heavy ion physics , dubna , 1986 , d7 - 87 - 68 ( dubna , 1987 ) , p. 528
; in proceedings of the 6th international conference on nuclear reaction mechanisms , varenna , 1991 , edited by e. gadioli ( ricerca scientifica ed educazione permanente supplemento n.84 , 1991 ) , p.39 . on heavy ion physics itkis m g _ et al .
_ 2002 proc . of the international conference on nuclear physics at border lines , may 21 - 24 , 2001 lipari ( messina ) , italy , eds .
g.fazio , g.giardina , f.hanappe , g.imme , n.rowley .
world scientific , singapore , p.146 - 156 fazio g _
et al . _ 2003 proc . of the international symposium on new projects and lines of research in nuclear physics ,
october 24 - 26 , 2002 , messina , italy , eds .
g.fazio , f.hanappe .
world scientific , singapore , p.258 - 271 | the fusion excitation function is the important quantity in planning experiments for the synthesis of superheavy elements .
its values seem to be determined by the experimental study of the hindrance to complete fusion by the observation of mass , angular and energy distributions of the fissionlike fragments .
there is ambiguity in establishment of the reaction mechanism leading to the observed binary fissionlike fragments .
the fissionlike fragments can be produced in the quasifission , fast fission , and fusion - fission processes which have overlapping in the mass ( angular , kinetic energy ) distributions of fragments .
the branching ratio between quasifission and complete fusion strongly depends on the characteristics of the entrance channel . in this paper
we consider a wide set of reactions ( with different mass asymmetry and mass symmetry parameters ) with the aim to explain the role played by many quantities on the reaction mechanisms .
we also present the results of study of the @xmath0ca+@xmath1bk reaction used to synthesize superheavy nuclei with z = 117 by the determination of the evaporation residue cross sections and the effective fission barriers @xmath2 of excited nuclei formed along the de - excitation cascade of the compound nucleus . |
the work has been supported by the european community - access to research infrastructure action of the improving human potential programme , by the daad exchange programme ( ppp - polen ) , by the polish state committe for scientific research ( grants no .
2p03b07123 and pb1060/p03/2004/26 ) and by the research centre jlich . | due to the high sensitivity of the @xmath0 reaction to the nucleon nucleon potential , bremsstrahlung radiation is used as a tool to investigate details of the nucleon nucleon interaction .
such investigations can be performed at the cooler synchrotron cosy in the research centre jlich , by dint of the cosy11 detection system .
+ the results of the identification of bremsstrahlung radiation emitted via the @xmath1 reaction in data taken with a proton target and a deuteron beam are presented and discussed .
+ the installation of a neutron detector at the cosy-11 facility [ 1,2 ] enables to study a plethora of new reaction channels .
it opens wide possibilities not only to investigate the isospin dependence of the meson production [ 3 ] , but also to measure the bremsstrahlung radiation created in the collisions of nucleons .
the study of the latter process is interested since it is highly sensitive to the kind of the nucleon - nucleon potential , and hence may serve as a tool to discriminate between various existing potential models [ 4,5 ] .
although bremsstrahlung radiation has been studied since many years , it is still the subject of interest of many theoretical and experimental groups [ 4,5,6,7 ] .
+ at the cosy11 experiment a signal from @xmath2quanta was observed in the time of flight distribution for the neutral particles measured between the target and the neutral particle detector [ 2 ] .
this encouraged us to analyse the data in view of the bremsstrahlung radiation in a free @xmath3 and a quasi
free @xmath4 reactions .
data have been taken using a proton target and a deuteron beam with a momentum close to the threshold of the @xmath5 process .
events corresponding to the @xmath3 and @xmath6 reaction have been identified by measuring the outgoing charged as well as neutral ejectiles .
the protons and deuterons are detected by means of drift chambers and scintillator hodoscopes while neutrons and photons are registered in a scintillator
lead sandwich type detector .
the momentum vectors of the charged ejectiles are reconstructed by tracking back the trajectories to the target point [ 1 ] . in case of a neutron the time of flight between the target and the neutral particle detector together with the known position of the hitted detection unit enables to determine its four momentum vector [ 8 ] . + in order to identify the @xmath3 reaction events with two tracks in the drift chambers and a simultaneous signal in the neutron detector have been selected . in fig.1 ( left ) the squared mass of one particle is plotted versus the squared mass of the other registered particle .
based on this figure the measured reactions can be grouped according to the type of ejectiles .
thus reactions with two protons , proton and pion , proton and deuteron , and pion and deuteron can be clearly selected .
next the distribution of the time of flight between the target neutron detector was determined with requirement that one of the charged particles was identified as a proton and the other as a deuteron . due to the baryon number conservation , gamma quanta
are the only one possible source of a signal in a neutron detector .
indeed a clear peak around the time corresponding to the time of light of the light is visible ( see fig.1 right ) .
the gamma quanta may originate from bremsstrahlung reaction or from the decay of produced mesons eg . via the @xmath7 .
it is possible to distinguish between these hypothesis calculating the missing mass produced in the @xmath8 reaction .
fig.2 left shows the distribution of the squared missing mass as obtained for the @xmath8 reaction . a significant peak around 0 mev@xmath9/c@xmath10 the squared mass of a gamma quanta
constitute an evidence for events associated to the deuteron
proton bremsstrahlung .
in addition a broad structure at higher masses originating from two pions emmited from reaction @xmath11 or two gamma quanta from @xmath12 reaction is visible .
the extraction of the total cross section of the @xmath3 reaction requires the luminosity and acceptance determination , which will be performed in the near future .
the analysis of the @xmath13 reaction is more complicated , however in this case all three baryons , namely two protons and neutron can be measured in the drift chambers and in the neutron detector , respectively .
fig.2 ( right ) presents the time of flight with the condition that in coincidence with a neutral particle also two protons were identified . a clear signal originating from gamma quanta
is seen at a time value of 24 ns which coresponds to the velocity of light . using the missing mass technique , it might be possible to verify if gamma quanta originate from the direct @xmath13 reaction . at present
the data analysis is in progress . |
the bean s critical state model@xcite successfully describes the irreversible magnetization in type - ii superconductors by introducing the critical current density @xmath2 , where @xmath3 is the temperature and @xmath4 is the magnetic field . in the framework of the bean s model
the value of the slope of the stationary magnetic field profile is less or equal to @xmath5 .
this nonuniform flux distribution does not correspond to an equilibrium state and under certain conditions flux jumps arise in the critical state .
the flux jumping process results in a flux redistribution towards the equilibrium state and is accompanied by a strong heating of the superconductor .
flux jumping has been numerously studied in conventional and high - temperature superconductors ( see the review papers @xcite , references therein , and the recent experimental studies @xcite ) . in the general case
two types of flux jumps can be considered , namely , the global and the local flux jumps .
a global flux jump involves vortices into motion in the entire volume of the sample .
a local flux jump happens in a small fraction of the sample volume .
depending on the initial perturbation and the driving parameters there are two qualitatively different types of global flux jumps , namely , complete and partial flux jumps .
the first turns the superconductor to the normal state .
the second self - terminates when the temperature is still less than the critical temperature .
we illustrate a global flux jump origination treating a superconducting slab with the thickness @xmath6 subjected to an external magnetic field @xmath7 parallel to the sample surface ( @xmath8 plane ) . in the framework of the bean s critical state model the space distribution of the flux
is given by the equation @xmath9 where the @xmath10 stays for @xmath11 and @xmath12 correspondingly .
we show the dependence @xmath13 in fig .
[ f1 ] for the case , when the critical current density depends only on the temperature , _
i.e. _ , @xmath14 .
let us now suppose that the temperature of the sample @xmath15 is increased by a small perturbation @xmath16 arising due to a certain initial heat release @xmath17 .
the critical current density @xmath18 is a decreasing function of temperature .
thus , the density of the superconducting current screening of the external magnetic field at @xmath19 is less than at @xmath20 .
this reduction of the screening current leads to a rise of the magnetic flux inside the superconductor as it is shown in fig .
the motion of the magnetic flux into the sample , which occurs as a result of the temperature perturbation @xmath16 , induces an electric field perturbation @xmath21 .
the arise of @xmath21 is accompanied by an additional heat release @xmath22 , an additional temperature rise @xmath23 , and , therefore , an additional reduction of the superconducting screening current density @xmath24 . under certain conditions
it results in an avalanche - type increase of the temperature and the magnetic flux in the superconductor , _
i.e. _ , in a global flux jump . the relative effect of the flux and temperature redistribution dynamics on the flux jumping process depends on the ratio , @xmath25 , of the flux , @xmath26 , and thermal , @xmath27 , diffusion time constants @xcite .
the value of the dimensionless parameter @xmath25 is determined by the corresponding diffusion coefficients and is equal to @xmath28 where @xmath29 is the heat conductivity , @xmath30 is the conductivity , and @xmath31 is the heat capacity . for @xmath32 ( @xmath33 ) , rapid propagation of the flux is accompanied by an adiabatic heating of the superconductor , _
i.e. _ , there is not enough time to redistribute and remove the heat released due to the flux motion .
for @xmath34 ( @xmath35 ) , the space distribution of flux remains fixed during the stage of rapid heating .
these adiabatic ( @xmath32 ) and dynamic ( @xmath34 ) approximations are the basis of the approach to the flux jumping problem @xcite and a flux jump scenario significantly depends on the relation between the values of the heat conductivity @xmath36 , heat capacity @xmath31 , and conductivity @xmath30 , that is determined by the slope of the @xmath37-@xmath38 curve .
let us now estimate the electric field value typical for the magnetization experiments . in this case
the external magnetic field ramp rate @xmath1 is usually from the interval
@xmath39 ts@xmath40 . the background electric field , @xmath41 , induced by the magnetic field variation is of the order of @xmath42 , where @xmath43 is the width of the area occupied by the critical state ( see , for example , fig . [ f1 ] ) .
we estimate @xmath41 as @xmath44 vcm@xmath40 using the value @xmath46 m which is typical for the stability domain of the bean s critical state .
this electric field interval corresponds to the flux creep regime .
therefore , for the magnetization experiments the background electric field @xmath41 is from the flux creep regime , where the relation between the current density , @xmath37 , and the electric field , @xmath38 , is strongly nonlinear . as a result
, the value of @xmath30 , _ i.e. _ , the slope of the @xmath37-@xmath38 curve , is strongly electric field dependent and the flux jumping takes place on a background of a resistive state with a conductivity that strongly depends on the external magnetic field ramp rate @xmath1 . in order to calculate the conductivity in the flux creep regime we use the dependence of @xmath37 on @xmath38 in the form @xmath47 where @xmath48 is the voltage criterion at which the critical current density @xmath24 is defined , @xmath49 determines the slope of the @xmath37-@xmath38 curve and @xmath50 .
note , that the actual choice of @xmath48 is not very essential .
indeed , by taking for the voltage criterion a certain value @xmath51 instead of @xmath48 we change the critical current density from @xmath24 to @xmath52 .
the difference between @xmath53 and @xmath24 is small as @xmath54 and @xmath50 .
it is common to define the critical current value as the current density at @xmath55 vcm@xmath40 .
let us also note , that a power law @xmath56 with @xmath57 is often used to describe the @xmath37-@xmath38 curve in the flux creep regime .
expanding the dependence given by eq .
( [ e3a ] ) in series in @xmath58 and keeping the first two terms we find that if we take @xmath59 , then eqs . ( [ e3 ] ) and ( [ e3a ] )
coincide with the accuracy of @xmath60 .
the relation given by eq .
( [ e3 ] ) was first derived in the framework of the anderson - kim model @xcite considering the thermally activated uncorrelated hopping of bundles of vortices .
the vortex - glass @xcite and collective creep @xcite models result in more sophisticated dependencies of @xmath37 on @xmath38 .
however , these @xmath37-@xmath38 curves coincide with the one given by eq .
( [ e3 ] ) if @xmath61 .
the recently developed self - organized criticality approach to the critical state @xcite also results in eq .
( [ e3 ] ) if @xmath61 .
the logarithmic dependence of the current density @xmath37 on the electric field @xmath38 in the interval @xmath61 is in a good agreement with numerous experimental data @xcite . in this paper
we use the @xmath37-@xmath38 curve given by eq .
( [ e3 ] ) to calculate the conductivity @xmath30 assuming that @xmath62 .
it follows from eq .
( [ e3 ] ) that for the flux creep regime the conductivity @xmath30 is given by the formula @xmath63 we estimate the value of @xmath30 as @xmath64 @xmath65cm@xmath40 using the typical data @xmath66 a@xmath67 and @xmath68 vcm@xmath40 .
it follows from this estimation that the conductivity @xmath30 determining the flux jumps dynamics for the magnetization experiments is very high . as a consequence
the dimensionless ratio @xmath25 is also very high .
thus , the scenario of a flux jump for the magnetization experiments corresponds to the limiting case when @xmath34 and the rapid heating stage takes place on the background of a `` frozen - in '' magnetic flux . the electric field dependent conductivity @xmath69 significantly affects the flux jumping process
. in particular , it results in the flux jump field @xmath0 dependence on the magnetic field ramp rate @xmath1 .
this dependence is known from experiments @xcite and was never considered theoretically as a consequence of the logarithmic @xmath37-@xmath38 curve characterizing the flux creep regime in superconductors with high values of the critical current density @xmath24 . under certain conditions
a flux jump is preceded by a series of magneto - thermal oscillations @xcite .
these oscillations have been observed for the low - temperature @xcite as well as for the high - temperature superconductors @xcite .
theoretically magneto - thermal oscillations were considered for a flux jump developing in the flux flow regime @xcite . in this case
the @xmath37-@xmath38 curve is linear and the value of the conductivity @xmath30 is electric field independent . the high and electric field dependent conductivity @xmath69 significantly affects the flux dynamics and therefore the magneto - thermal oscillations . in particular
, it results in the dependence of the frequency of the magneto - thermal oscillations on the magnetic field ramp rate @xmath1 .
the effect of the logarithmic @xmath37-@xmath38 curve on the magneto - thermal oscillations was never treated theoretically . in this paper
we consider the flux jump instability of the bean s critical state arising in the flux creep regime on the background of a nonuniform electric field determining the conductivity of the type - ii superconductor in the flux creep regime .
we find the flux jump field , @xmath0 , that determines the critical state stability criterion and the dependence of @xmath0 on the external magnetic field , @xmath70 , and the external magnetic field ramp rate , @xmath1 .
we show that a flux jump can be preceded by the magneto - thermal oscillations and find the frequency of these oscillations as a function of @xmath1 .
the paper is organized in the following way . in sec .
ii , we consider the critical state stability qualitatively and obtain the stability criterion . in sec .
iii , we derive the equations determining the development of the small temperature and electric field perturbations and calculate the frequency of the magneto - thermal oscillations . in sec .
iv , we summarize the overall conclusions .
in this section we consider the critical state stability qualitatively assuming that the thermomagnetic instability develops much faster than the magnetic flux diffusion process .
in other words , we treat the case when the heating accompanying the thermomagnetic instability takes place on the background of a `` frozen - in '' magnetic flux . in sec .
iii we derive the exact criterion of applicability of the following qualitative reasoning .
let us consider a superconducting slab with the thickness @xmath6 subjected to a magnetic field parallel to the sample surface ( see fig . [ f1 ] ) and suppose that the temperature of the sample @xmath15 is increased by a small perturbation @xmath71 . to keep the critical state stable , _
i.e. _ , to keep the screening current at the same level , an electric field perturbation @xmath72 arises .
the additional electric field @xmath72 causes an additional heat release @xmath73 , which is the `` price '' for keeping the total screening current density at the same level , _
i.e. _ , the `` price '' for the `` frozen - in '' magnetic flux .
the critical state is stable if the additional heat release @xmath74 can be removed to the coolant by the additional heat flux @xmath75 resulting from the temperature perturbation @xmath71 .
thus , the critical state stability criterion has the form @xmath76 the additional heat release per unit length , @xmath74 , is given by the integral of @xmath77 over the width of the superconducting slab @xmath78 the additional heat flux , @xmath79 , is determined by the temperature perturbation @xmath71 at the sample surface , _ i.e. _ , @xmath80 where @xmath81 is the heat transfer coefficient to the coolant with the temperature @xmath15 and @xmath82 stays for the sample surface . using eqs .
( [ e5 ] ) , ( [ e6 ] ) and ( [ e7 ] ) we find the critical state stability criterion , namely , the inequality @xmath83 to derive the explicit form of this stability criterion we have to find the relation between @xmath71 and @xmath72 . to do it ,
we calculate the decrease of the current density @xmath84 resulting from the temperature perturbation @xmath71 and the increase of the current density @xmath85 resulting from the electric field perturbation @xmath72 .
if the critical state is stable then the total screening current density stays constant . as a result , the relation between @xmath72 and @xmath71 is given by the equation @xmath86 in the critical state , @xmath87 , thus , the decrease of @xmath37 due to the temperature perturbation @xmath71 is equal to @xmath88 ( note that @xmath89 ) .
the increase of the current density due to the electric field perturbation @xmath72 can be written as @xmath90 note , that the conductivity @xmath30 is the differential conductivity , _
i.e. _ , it is determined by the slope of the @xmath37-@xmath38 curve .
combining eqs .
( [ e4 ] ) and ( [ e11 ] ) , we find the relation between @xmath85 and @xmath72 in the form @xmath91 where @xmath92 .
it follows from eqs .
( [ e9 ] ) , ( [ e10 ] ) and ( [ e12 ] ) that @xmath93 equations ( [ e4 ] ) and ( [ e13 ] ) allow to understand the effect of the background electric field @xmath41 on the critical state stability .
it follows from eq .
( [ e4 ] ) that a low electric field @xmath41 results in a high differential conductivity ( @xmath94 ) . in its turn
a high conductivity @xmath30 leads to a low electric field perturbation ( indeed , it follows from eq .
( [ e13 ] ) that @xmath95 ) .
the smaller is @xmath72 , the less `` costly '' it is to remove the additional heat release . as a result the lower is the background electric field @xmath41 the more stable is the superconducting state . substituting eq .
( [ e13 ] ) into eq .
( [ e8 ] ) we find the critical state stability criterion in the form @xmath96 we have to treat the temperature perturbation @xmath71 in more detail to derive the final form of eq .
( [ e14 ] ) .
the variation of the function @xmath97 on the interval @xmath98 depends on the value of the biot number @xmath99 where @xmath36 is the heat conductivity of the superconductor .
let us assume that the value of the heat transfer coefficient @xmath81 is relatively low .
as a result , @xmath100 and the temperature perturbation @xmath97 is almost uniform over the width of the superconducting slab .
it means that @xmath71 cancels in both sides of eq .
( [ e14 ] ) and the bean s critical state stability criterion takes the following final form @xmath101 let us note , that this criterion was first derived in order to calculate the maximum value of a superconducting current under conditions typical for the critical current measurements , _
i.e. _ , for a superconducting wire carrying a current that is increased with a given ramp rate @xcite .
in addition , we assume for simplicity that the value of @xmath102 is temperature and magnetic field independent if @xmath103 and @xmath104 , where @xmath105 is the critical temperature and @xmath106 is the upper critical field .
this assumption is in a good agreement with numerous experimental data @xcite as well as with the self - organized criticality approach to the bean s critical state @xcite . using eq .
( [ e1 ] ) we can rewrite the criterion given by eq .
( [ e16 ] ) in the following form which is convenient for the further analysis @xmath107 where @xmath108 is the magnetic field in the middle plane of the superconducting slab .
the background electric field @xmath41 is induced by the varying external magnetic field @xmath109 and thus the spatial distribution of @xmath41 is given by the maxwell equation @xmath110 combining eqs .
( [ e1 ] ) and ( [ e18 ] ) and taking into account that @xmath87 we find that @xmath111 where the @xmath10 stays for @xmath11 and @xmath12 correspondingly . at the same time eq .
( [ e18 ] ) results in the relation @xmath112 it follows from eqs .
( [ e19 ] ) and ( [ e20 ] ) that the the background electric field @xmath41 dependence on @xmath4 is given by the formula @xmath113 where the @xmath10 stays for @xmath11 and @xmath12 correspondingly .
let us now apply the criterion given by eq .
( [ e17 ] ) to calculate the flux jump field @xmath0 assuming that initially there is no flux inside the superconducting slab , _
i.e. _ , we calculate now the magnetic field of the first flux jump . using eqs .
( [ e17 ] ) and ( [ e21 ] ) we find the stability criterion in the form @xmath114 the value of the magnetic field @xmath115 is given by the following system of equations @xmath116 where the value of the magnetic flux penetration field , @xmath117 , is determined by @xmath118 it follows from eqs .
( [ e22 ] ) and ( [ e24 ] ) that @xmath119 is an increasing function of the external magnetic field @xmath70 if @xmath120 and @xmath119 is a decreasing function of @xmath70 if @xmath121 .
in other words , if for a given value of @xmath1 the superconducting state is stable in the region @xmath122 then it is stable for any magnetic field .
thus , if a flux jump occurs it occurs only if @xmath120 .
therefore , we consider now a superconducting slab that is wide enough meaning that @xmath123 .
we have the criterion @xmath124 to find the flux jump field @xmath0 in the case when @xmath120 .
thus , it follows from eq .
( [ e22 ] ) that the dependence @xmath125 is given by the equation @xmath126 let us approximate the value of @xmath127 as @xmath128 using eq .
( [ e27 ] ) we rewrite eq .
( [ e26 ] ) in the following form @xmath129 we treat now the case when @xmath130 or in other words @xmath131 .
it means that @xmath132 , where @xmath105 is the critical temperature at zero magnetic field .
it follows finally from eq .
( [ e28 ] ) that the bean s critical state stability criterion determining the the dependence @xmath125 is given by the equation @xmath133 let us now consider the particular case of the bean s critical state model , namely , let us assume that the critical current density is magnetic field independent , _
i.e. _ , @xmath134 using eq .
( [ e29 ] ) we find the following formula for the first flux jump field @xmath0 @xmath135 it follows from eq .
( [ e31 ] ) that the value of @xmath0 is inversely proportional to the square root of the magnetic field ramp rate @xmath1 and is , therefore , decreasing with the increase of @xmath1 .
the physics of this effect is related to the decrease of the conductivity @xmath69 in the flux creep regime with the increase of the background electric field @xmath41 , _ i.e. _ , with the increase of @xmath1 .
we derive the expression for @xmath0 assuming that the rapid heating stage of a flux jump takes place on the background of a `` frozen - in '' magnetic flux .
this approach is valid if @xmath34 which is the same as @xmath136 where we introduce the typical thermal diffusion time constant @xmath137 as @xmath138 let us now compare the values of @xmath0 and @xmath139 , where @xmath139 determines the flux jump field for the adiabatic stability criterion @xcite . this well known criterion is based on the suggestion that the heating accompanying a flux jump is an adiabatic process , _ i.e. _ , it is assumed that there is no heat redistribution during a flux jump .
therefore , the adiabatic stability criterion corresponds to the limiting case of @xmath32 that is not the case typical for the magnetization experiments .
the value of @xmath139 is given by the formula @xcite @xmath140 it follows from the comparison of eqs .
( [ e31 ] ) and ( [ e32 ] ) that @xmath141 if @xmath142 note , that the inequality given by eq .
( [ e33 ] ) is stronger than the one given by eq .
( [ e35 ] ) as we assume that @xmath100 .
the critical current density is decreasing with the increase of the magnetic field .
let us now consider the effect of this dependence on the critical state stability assuming that the value of @xmath0 is relatively high . to do it we use the kim - anderson model @xcite to describe the function @xmath143 . in the case of a high magnetic field
this model results in the relation @xmath144 using eqs .
( [ e36 ] ) and ( [ e29 ] ) we find for the flux jump field @xmath0 the formula @xmath145 the comparison of eqs .
( [ e37 ] ) and ( [ e31 ] ) shows that the magnetic field dependence of the critical current density slows down the decrease of @xmath0 with the increase of @xmath1 .
we treat now the critical state stability in more detail and , in particular , we take into consideration the magneto - thermal oscillations . we consider a superconducting slab with the thickness @xmath6 subjected to an external magnetic field parallel to the @xmath146 axis ( see fig . [ f1 ] ) .
we use for calculation the bean s critical state model assuming that the critical current density is magnetic field independent , _
i.e. _ , @xmath14 .
we suppose also that @xmath147 .
the space distribution of the background electric field @xmath41 is then given by the formulae @xmath148 where the magnetic field penetration depth , l , is equal to @xmath149 we consider now the stability of the stationary electric field and temperature distributions corresponding to the bean s critical state against small electric field and temperature perturbations .
to do it let us present the electric field @xmath150 and the temperature @xmath151 in the following form @xmath152 @xmath153 where @xmath154 @xmath155 , @xmath156 , @xmath157 is the characteristic time of the magneto - thermal instability increase , and @xmath158 is the frequency of the magneto - thermal oscillations .
the stationary temperature @xmath159 is different from @xmath15 due to the joule heating power @xmath160 produced by the background electric field @xmath41 .
let us note , that the difference between @xmath159 and @xmath15 is small , _
i.e. _ , @xmath161 . indeed , using eq .
( [ e29 ] ) we estimate the value of @xmath162 as @xmath163 .
the small temperature @xmath164 and electric field @xmath165 perturbations decay if @xmath166 .
therefore , the stability margin of the bean s critical state is determined by the condition @xmath166 . substituting eqs .
( [ e40 ] ) and ( [ e41 ] ) into the maxwell equation and the heat diffusion equation @xmath167 we find the system of equations describing @xmath168 and @xmath169 .
it takes the form @xmath170 where the prime is to denote the derivative over x. note , that to derive eq .
( [ e45 ] ) we use eq .
( [ e38 ] ) and the relation @xmath171 we assume that the superconducting slab is in a thermal contact with a coolant with the temperature @xmath15 and we treat here the case when the external magnetic field ramp rate is given , _
i.e. _ , @xmath172 .
in addition , the electric field @xmath173 is equal to zero in the inner region of the superconducting slab ( @xmath174 ) . as a result , the boundary conditions for eqs .
( [ e44 ] ) and ( [ e45 ] ) are given by @xmath175 @xmath176 let us present the solution for @xmath168 in the form @xmath177 where the first term corresponds to the approximation of a `` frozen - in '' magnetic flux ( see eq .
( [ e13 ] ) ) and the second term describes the deviation from this approximation .
it follows from eqs .
( [ e45 ] ) , ( [ e48 ] ) , and ( [ e49 ] ) that with the accuracy of @xmath100 the equation for @xmath178 has the form @xmath179 with the boundary conditions @xmath180 we consider here the case of @xmath34 , _
i.e. _ , the case when in the first approximation in @xmath181 the magnetic flux is `` frozen - in '' in the bulk of the superconducting slab .
it means that during the rapid heating stage the magnetic flux redistribution takes place only in a thin surface layer with the thickness @xmath182 . in other words
, the function @xmath178 decays inside the superconducting slab and differs from zero only if @xmath183 is less or of the order of the skin depth @xmath184 . in the region
@xmath185 eq .
( [ e50 ] ) takes the form @xmath186 the solution of eq .
( [ e52 ] ) matching the boundary conditions given by eq .
( [ e51 ] ) reads @xmath187 where we introduce the value of the skin depth @xmath184 as @xmath188 to find the values of @xmath189 and @xmath158 we integrate now eq .
( [ e44 ] ) over @xmath190 from @xmath191 to @xmath192 .
using eqs .
( [ e47 ] ) , ( [ e49 ] ) , and ( [ e53 ] ) we find the equation determining @xmath193 in the form : @xmath194 we show schematically the dependencies of @xmath189 and @xmath158 on @xmath70 in fig . [ f2 ] , where the field @xmath195 is determined by the equation @xmath196 the difference between @xmath195 and @xmath0 is small in the case when the ramp rate @xmath1 is low , _
i.e. _ , @xmath197 if @xmath198 it follows from eq .
( [ e55 ] ) that @xmath199 if @xmath200 , _ i.e. _ , the bean s critical state is stable if @xmath201 , where the flux jump field @xmath0 is given by the eq .
( [ e31 ] ) .
we find also that at the stability threshold ( for @xmath200 ) the value of @xmath193 is imaginary , _ i.e. _ , @xmath202 .
thus the magneto - thermal instability is preceded by the magneto - thermal oscillations with the frequency , @xmath203 , given by the formula @xmath204 the `` frozen - in '' magnetic flux approximation is valid if the surface layer where @xmath205 is thin , _
i.e. _ , if @xmath182 .
using eqs .
( [ e32 ] ) , ( [ e39 ] ) , ( [ e54 ] ) , and ( [ e56 ] ) we find the applicability criterion of the above approach in the form @xmath206
to summarize , we consider the flux jump instability of the bean s critical state in type - ii superconductors . we show that under the conditions typical for most of the magnetization experiments this instability arise in the flux creep regime .
we find the flux jump field @xmath0 that determines the critical state stability criterion .
we show that the bean s critical state stability is determined by the slope of the current - voltage curve .
we calculate the dependence of @xmath0 on the external magnetic field ramp rate @xmath1 .
we find the frequency of the magneto - thermal oscillations preceding a flux jump as a function on the external magnetic field ramp rate @xmath1 .
bean , , 250 ( 1962 ) ; , 31 ( 1964 ) .
mints and a.l .
rakhmanov , , 551 ( 1981 ) .
wipf , cryogenics * 31 * , 936 ( 1991 ) .
mchenry , h.s .
lessure , m.p .
maley , j.y .
coulter , i. tanaka , and h. kojima , physica * c 190 * , 403 ( 1992 ) .
l. legrand , i. rosenman , ch .
simon , and g. collin , physica * c 211 * , 239 ( 1993 ) .
a. gerber , j.n .
li , z. tarnawski , j.j.m .
franse , and a.a .
menovsky , , 6047 ( 1993 ) .
p. leiderer , j. boneberg , p. br " ull , v. bujok , and s. herminghaus , , 2646 ( 1993 ) .
anderson , , 309 ( 1962 ) .
kim , c. hempstead , and a. strnad , , 306 ( 1962 ) ; phys .
rev . * 131 * , 2484 ( 1963 ) .
anderson and y.b .
kim , , 39 ( 1964 ) .
fisher , , 1415 ( 1989 ) ; d.s .
fisher , m.p.a .
fisher , and d.a .
huse , , 130 ( 1991 ) .
feigelman , v.b .
geshkenbein , a.i .
larkin , and v.m .
vinokur , , 2303 ( 1989 ) .
t. nattermann , , 2454 ( 1990 ) ; k.h .
fisher and t. nattermann , , 10372 ( 1991 ) .
z. wang and d. shi , solid state commun .
* 90 * , 405 ( 1994 ) .
w. pan and s. doniach , , 1192 ( 1994 ) .
see a. gurevich and h. kpfer , , 6477 ( 1993 ) and the references therin . n.h .
zebouni , a. venkataram , g.n .
rao , c.g .
grenier , and j.m .
reynolds , phys .
* 13 * , 606 ( 1964 ) .
j. chikaba , cryogenics * 10 * , 306 ( 1970 ) .
mints , jetp lett .
* 27 * , 417 ( 1978 ) .
mints and a.l .
rakhmanov , j. phys .
d : appl .
phys . * 15 * , 2297 , 1982 .
r. hancox , phys .
* 16 * , 208 ( 1965 ) . | we consider the flux jump instability of the bean s critical state arising in the flux creep regime in type - ii superconductors .
we find the flux jump field , @xmath0 , that determines the superconducting state stability criterion .
we calculate the dependence of @xmath0 on the external magnetic field ramp rate , @xmath1 .
we demonstrate that under the conditions typical for most of the magnetization experiments the slope of the current - voltage curve in the flux creep regime determines the stability of the bean s critical state , _
i.e. _ , the value of @xmath0 . we show that a flux jump can be preceded by the magneto - thermal oscillations and find the frequency of these oscillations as a function of @xmath1 . |
spin physics is a rapidly growing area of research in condensed matter science aimed at the creation , manipulation and detection of spins in various systems .
important and interesting fundamental results , also promising for possible future applications , have been obtained in semiconductors and semiconductor nanostructures @xcite .
highly sensitive methods of kerr and faraday rotation spectroscopy give a possibility to probe spin orientation within subpicosecond resolution @xcite . a challenging problem in the spin physics
is how to affect the spin by instantaneous non - magnetic methods , in particular , by static or optical electric fields . in semiconductors ,
optical selection rules allow one to create spin - oriented carriers ( optical orientation ) which may then be accelerated by a static external electric field .
another possibility to monitor the spin by electrical means arises due to the spin - orbit interaction which couples the electron orbital and spin degrees of freedom . in semiconductor heterostructures ,
the spin - orbit interaction is described by the rashba and dresselhaus contributions to the electron effective hamiltonian linear in the electron quasimomentum @xmath0 , for review see ref .
generally , they can be presented as follows @xmath1 where @xmath2 ( @xmath3 ) are the pauli matrices , @xmath4 is a second - rank pseudotensor , and the effective momentum - dependent larmor frequency is defined by @xmath5 .
hereafter , for definiteness , we consider zinc - blende - lattice based quantum wells grown along the direction [ 001 ] and use the coordinate frame @xmath6 , y \parallel [ 110]$ ] and @xmath7 $ ] .
we study here the electron spin dynamics in semiconductor heterostructures in the presence of an external in - plane electric field .
the attention is focused on the so - called streaming regime @xcite characterized by the periodic optical - phonon emission by electrons subjected to acceleration in the strong electric field .
a number of interesting time - resolved spin effects are shown to be caused by the electric field .
we also discuss a possibility to detect them in the pump - probe experiments , both in the absence and presence of an external magnetic field .
in an external electric field @xmath8 , the carrier distribution is anisotropic in the momentum space . in particular , it is characterized by a nonzero electron drift momentum @xmath9 in the opposite to field direction .
for comparatively weak fields @xmath10
kv / cm , the anisotropy is weak : the ratio @xmath11 is smaller than @xmath12 , where @xmath13 is the root - mean - square of the electron quasimomentum in the absence of the field .
an application of the increasing field leads to a transformation from the linear dependence @xmath14 to the saturated behaviour .
if the interparticle collisions play a minor role in the kinetics , the electron momentum - space distribution in strong electric fields becomes extremely anisotropic , _
each electron accelerates quasiballistically in the `` passive '' region until reaching the threshold energy equal the optical - phonon energy @xmath15 and the corresponding threshold quasimomentum @xmath16 , where @xmath17 is the electron effective mass .
then it looses its energy by emitting an optical phonon and starts the next period of acceleration .
thus , in the streaming regime realized in an appropriate range of the dc electric fields , the electron distribution in the momentum space has a steaming - like , or needle - like , form spread between the @xmath18-point @xmath19 and the point @xmath20 , where @xmath21 is the unit vector in the direction opposite to the electric field @xmath22 .
the formation of such an anisotropic electron distribution requires the following hierarchy of times @xmath23 here @xmath24 is the time of optical phonon emission by an electron in the `` active '' energy region exceeding @xmath15 , @xmath25 is the travel time through the passive region of the momentum space , @xmath26 is the momentum relaxation time due to electron scattering by acoustic phonons or static imperfections , and @xmath27 is the electron - electron collision time .
the electron - electron scattering tends to convert the anisotropic distribution to the shifted maxwellian distribution with an effective electron temperature and a drift velocity @xmath28 .
the corresponding time can be estimated by @xmath29 , where @xmath30 is the electron 2d density and @xmath31 is the dielectric constant . the electron distribution function @xmath32 is given by @xcite @xmath33 where @xmath34 , @xmath35 , and @xmath36\:.\ ] ] one can see from eq .
( [ spin_independ ] ) that the penetration @xmath37 into the active region and the width @xmath38 of the distribution in the transverse direction @xmath39 are given by @xmath40 and @xmath41 .
thus , the needle - like distribution is formed provided @xmath42 which establishes the upper limit for the electric field strength . in this case
the function @xmath43 can be approximated by the heaviside function @xmath44 and @xmath45 by @xmath46 .
it is also worth to mention that the value @xmath47 gives the ratio between the numbers of particles in the passive and active regions as well as the ratio @xmath48 of the times spend by electrons in these regions .
if the electrons are spin polarized with the average spin @xmath49 per particle , their spin distribution function in the @xmath50-space is given by the main contribution @xmath51 and an additional term of the order of @xmath52 . in order to analyze consequences of elastic scattering by static imperfections , impurities and interface roughness , and/or quasi - elastic scattering by acoustic phonons , we start with the approximation of `` ideal streaming '' replacing the smoothed delta - function @xmath53 in eq .
( [ spin_independ ] ) by the dirac delta - function and the smoothed theta - function @xmath54 by the heaviside step function .
then the streaming distribution in the passive region reduces to @xmath55 in the following we include into consideration the elastic scattering by short - range imperfections and describe the scattering process @xmath56 by the rate @xmath57 normalized on the inverse momentum relaxation time @xmath26 as follows @xmath58 since the streaming regime is achieved provided @xmath26 is much longer than the travel time @xmath59 we treat the elastic scattering keeping the first order corrections in @xmath60 to the electron distribution function ( [ zero ] ) .
then the scatter of electrons over all azymuthal angles @xmath61 gives rise to a correction @xmath62 which , in the passive region , satisfies the equation @xmath63 with the boundary condition @xmath64 at @xmath65 . here
the right - hand side term represents the generation of electrons to the whole area @xmath66 due to the elastic scattering from the zeroth - approximation needle ( [ spin_independ ] ) while the term on the left - hand side describes the further acceleration of the scattered electrons in the electric field .
substituting eq . into eq .
( [ first ] ) we obtain @xmath67 the solution of this first - order differential equation leads to the formula @xmath68 the steady - state density of electrons scattered off the needle in the passive area is given by @xmath69 } = n { t_{\rm tr}\over \tau_p } { 2g+1\over \pi } \approx 0.9 \ , n { t_{\rm tr}\over \tau_p}\:,\ ] ] where @xmath70 is the catalan constant .
these electrons reach the sphere @xmath71 , emit optical phonons and return to the point @xmath72 thus , in the first approximation , the total density @xmath30 should be replaced by @xmath73 . the perturbation procedure can be continued by the replacements @xmath74 and @xmath75 in the right- and left - hand sides of eq .
( [ first ] ) , respectively .
the constriction of the distribution function @xmath76 along the @xmath39 axis and the electron homogeneous occupation of the @xmath77 interval between 0 and @xmath78 substantially modifies the dyakonov - perel spin relaxation of the @xmath79 component controlled by the linear-@xmath50 term @xmath80 in eq .
( [ h_so ] ) . here , as above
, we assume the electric field to be applied along the @xmath81 axis .
the spin - relaxation time @xmath82 of this component is found from the equation , see @xcite : @xmath83 where @xmath84 , @xmath49 is the average electron spin per particle , @xmath85 is the first - order correction ( in @xmath86 ) to the spin distribution function satisfying the spin kinetic equation , see eq . (
22 ) in ref .
@xcite , @xmath87 solving this equation with the streaming distribution we obtain @xcite @xmath88 we remind that , in the state of equilibrium , the relaxation time @xmath82 is given by the similar equation with @xmath89 replaced by @xmath90 , where @xmath91 is the equilibrium value of the electron average energy .
now we consider effect of elastic scattering on the spin relaxation and show that even the first - order correction in the ratio @xmath92 can modify @xmath82 .
for this purpose we replace @xmath76 in eq .
( [ deltas ] ) by @xmath62 and find the scattering - induced correction @xmath93 .
the latter is given therefore by the integral @xmath94 which can be solved analytically to give @xmath95.\ ] ] substituting this correction to eq .
( [ sytausy ] ) and summing over @xmath50 we arrive at the contribution of the elastically scattered electrons to spin relaxation of the electron gas @xmath96 in contrast to the electrons in the zeroth - approximation needle distribution ( [ spin_independ ] ) compressed along the @xmath39 axis , the scattered electrons are spread over the whole passive area .
the increased precession frequency @xmath97 can compensate smallness of the ratio @xmath60 . for comparison of the contributions ( [ tausy ] ) and ( [ elasttaus ] )
it is instructive to sum them and present the total spin relaxation time in the following convenient form @xmath98 for @xmath99 and @xmath100 , we find the contribution ( [ elasttaus ] ) caused by elastic scattering to exceed that given by eq .
( [ tausy ] ) by more than twice .
in systems with the effective hamiltonian containing the @xmath0-linear terms , the application of an electric field results in orientation of electron spins . such electrical spin orientation is allowed by the symmetry only in gyrotropic systems where some components of the second - rank pseudotensor @xmath4 are nonzero . in weak fields ,
this effect has been studied in many works , see the review @xcite , references therein , and the recent paper @xcite . the spin polarization created by a weak electric field
is given by @xmath101 where @xmath102 and @xmath13 are the zero - field root - mean - square electron velocity and momentum , and @xmath103 .
it is remarkable that the coefficient @xmath104 depends on the ratio of spin and energy relaxation times . at slow energy relaxation ,
the spin is created independently at each energy @xmath105 resulting in a nonequilibrium spin distribution @xmath106 , where @xmath107 is the fermi - dirac distribution .
in contrast , the fast energy relaxation establishes the equilibrium distribution within each spin - split subband , and the nonequilibrium spin distribution function @xmath108 .
the integration over the energy yields different results for the spin polarization @xmath79 . for a short - range momentum scattering potential , in two limiting cases of fast and slow energy relaxation the coefficient @xmath104 in eq
. differs by a factor of two @xcite : @xmath109 and @xmath110 .
the results for these two extreme cases are shown by straight lines 1 in fig .
[ fig_spin_vs_e ] .
one can see that the spin polarization in the fields @xmath111 v / cm does not exceed 0.1% .
mev , @xmath112 mev , @xmath113 , @xmath114 ps , @xmath115 ps , @xmath116 @xmath117 , and zero temperature is assumed . ] in strong fields where the streaming regime is realized , the situation changes drastically .
treating the term @xmath118 in eq .
as the spin - orbit splitting of the spectrum we get the shifted needle - like distributions in the spin subbands with the projection @xmath119 onto the @xmath81 axis . then taking into account the terms @xmath120 and @xmath121 as the components of the precession frequency and neglecting elastic scattering processes in the passive area
, we obtain the spin polarization in the form @xcite @xmath122 where @xmath123 and @xmath124 stands for the fresnel sine integral . for small values of @xmath125
the ratio @xmath126 is approximated by @xmath127 and eq .
( [ synew ] ) turns into @xmath128 .
it follows from eq . that the electrically oriented spin is a non - monotonous function of the field strength in the streaming regime .
the dependence is shown by curve 2 in fig .
[ fig_spin_vs_e ] .
the values of the spin orientation have an order of a few percent which is more than an order of magnitude larger than those at low fields .
we note that the decrease in the spin relaxation time caused by elastic scattering and described by eqs .
( [ elasttaus ] ) , ( [ totalsr ] ) is compensated by a corresponding increase in the generation rate .
consider a undoped 2d structure excited by a short interband pulse creating photoelectrons at the bottom of the lowest conduction subband . assuming a homogeneous pulse , the non - equilibrium distribution at the initial moment @xmath129 is given by @xmath130 where @xmath30 is the photoelectron density . in the ideal streaming - regime model , the kinetic evolution of the photoelectrons is also described by a product of two delta - functions @xmath131 with the variable of one function being a function of time with the period @xmath59 . here
the symbol @xmath132 means the fractional part of the number @xmath133 .
if the optical pulse is circularly polarized then the generated carriers are spin oriented , and the spin energy distribution is oscillating in time as well . neglecting the splitting of spin states and the spin relaxation we can supplement eq . ( [ tinitt ] ) by the similar equation for the time - dependent spin distribution function , cf .
@xmath134 where @xmath135 is the initial electron spin per particle .
the both functions ( [ tinitt ] ) and ( [ tinits ] ) are periodic until the elastic scattering is ignored . the simplest way to take into account the elastic scattering of the accelerating bunch of electrons
is just multiply the above time - dependent functions by the exponential function @xmath136 turning them into damped oscillatory functions .
the temporary oscillations of @xmath137 can be detected in the pump - probe time - resolved experiments based on the spin kerr or faraday effects .
for the probe light frequency slightly below the quantum - well fundamental edge @xmath138 , the kerr rotation angle of polarization plane of the linearly polarized probe pulse is given by @xmath139 where @xmath105 is the electron energy referred to the bottom of the lowest 2d conduction subband , the energy - dependent coefficient @xmath140 is proportional to @xmath141^{-1}$ ] , @xmath142 is the in - plane electron effective mass , and @xmath143 is the reduced mass of an electron and a heavy hole .
the substitution of the spin distribution ( [ tinits ] ) multiplied by @xmath136 into eq .
( [ kerr ] ) gives @xmath144 where @xmath145 is a correction arising due to the elastically scattered carriers and proportional to the ratio @xmath60 . while deriving eq .
( [ thetat ] ) we took into account that the electron energy in the bunch varies in time as @xmath146 . , @xmath147 calculated neglecting the spin - orbit splitting ( a ) at zero magnetic field , and ( b ) in the external magnetic field with @xmath148 which corresponds to @xmath149 t at @xmath150 .
, title="fig : " ] , @xmath147 calculated neglecting the spin - orbit splitting ( a ) at zero magnetic field , and ( b ) in the external magnetic field with @xmath148 which corresponds to @xmath149 t at @xmath150 . , title="fig : " ] in the external in - plane magnetic field @xmath151 the electron spin component @xmath152 rotates around the @xmath153 axis with the angular frequency @xmath154 , where @xmath155 is the electron @xmath155-factor and @xmath156 is the bohr magneton .
the kerr rotation angle is obtained from eq .
( [ thetat ] ) by multiplying the right - hand side by @xmath157 . indeed , in the absence of the spin - orbit splitting , the spin precession plays a role of the clock independent of the periodic acceleration and emission of optical phonons by the photoelectrons .
the time dependence of the kerr rotation signal at zero and nonzero magnetic field is plotted in fig .
[ fig_kerr ] , ( a ) and ( b ) respectively .
only the first main contribution in eq .
( [ thetat ] ) is taken into account in the calculation .
the curve in fig . [ fig_kerr](a ) has a period @xmath59 whereas the curve in fig .
[ fig_kerr](b ) has two oscillations with two different periods , the longer one equal to the larmor precession period @xmath158 and the shorter one equal to the travel time @xmath59 .
we consider again optical creation of the spin @xmath159 at the bottom of the conduction band at the moment @xmath129 . the time - dependent spin distribution is then given by @xmath160 taking into account the spin - orbit splitting we obtain that the spin dynamics in the absence of external magnetic fields is described by @xcite @xmath161 where @xmath162 is the initial spin , and the phase @xmath163 changes in time according to @xmath164 + \left\ { { t \over
t_{\rm tr}}\right\}^2 \biggr),\ ] ] with the symbol @xmath165 $ ] denoting the integer part of @xmath81
. here @xmath166 is the spin precession frequency in the electric field . in weak fields it is linear in @xmath167 @xcite while in the streaming regime one has @xcite @xmath168 if @xmath169 is a rational part of @xmath170 then the functions @xmath171 are periodic . in particular , for @xmath172 with an integer @xmath173 , the period equals to @xmath174 . for irrational values of @xmath175 ,
the variation @xmath176 is aperiodic .
the time dependence @xmath177 for these two cases is shown in fig .
[ fig_spin_beats_el_field ] .
equal to @xmath170 ( a ) and @xmath178 ( b ) ; ( lower panel , c ) the time - averaged spin @xmath179 at @xmath180 calculated in the streaming regime neglecting elastic scattering .
inset highlights the narrow region of electric fields marked by dashed lines.,title="fig : " ] equal to @xmath170 ( a ) and @xmath178 ( b ) ; ( lower panel , c ) the time - averaged spin @xmath179 at @xmath180 calculated in the streaming regime neglecting elastic scattering .
inset highlights the narrow region of electric fields marked by dashed lines.,title="fig : " ] + equal to @xmath170 ( a ) and @xmath178 ( b ) ; ( lower panel , c ) the time - averaged spin @xmath179 at @xmath180 calculated in the streaming regime neglecting elastic scattering .
inset highlights the narrow region of electric fields marked by dashed lines.,title="fig : " ] according to eq .
( [ phit ] ) the phase @xmath163 is a quadratic function of @xmath181 , this explains why in fig .
[ fig_spin_beats_el_field ] the variation of oscillatory function @xmath177 in positive and negative areas is not antisymmetric . for the average spin @xmath182
we find a sharp electric - field dependence which can be called `` mode - locking '' . introducing @xmath183 $ ] and @xmath184 , we obtain @xmath185 where @xmath186 the expression for @xmath187 has a similar form .
one can see that the average spin @xmath188 is nonzero if the following condition is fulfilled : @xmath189 in this case the spin makes @xmath173 full turns around the @xmath153 axis during the time @xmath59 , and the average spin polarization has a high value .
the oscillatory behaviour of @xmath177 in fig .
[ fig_spin_beats_el_field](a ) corresponds to @xmath190 . within the period for @xmath191 lying between @xmath192 and @xmath193 ( @xmath194 ) the spin @xmath195-component is decreasing from its initial value @xmath162 ,
reaches the minimum value @xmath196 and rapidly increases up to @xmath162 .
nonlinear variation of @xmath177 with @xmath181 results in a nonvanishing value of the average @xmath179 .
moreover , for @xmath197 , the average @xmath179 is still remarkable in the close proximity of @xmath198 to @xmath199 where @xmath200 . in the rest area of @xmath198 values
the @xmath188 have the order of magnitude of @xmath201 , see eq . .
the electric field dependence of @xmath179 for @xmath202 is shown in the lower panel of fig .
[ fig_spin_beats_el_field ] , the inset in figure represents the selected area around @xmath203 in the enlarged scale .
allowance for other mechanisms of electron scattering and spin relaxation leads to a decay of the oscillation amplitude .
figure [ fig_spin_beats_el_field_damp ] demonstrates the effect of elastic scattering on the dynamics of @xmath204 and @xmath205 components which is taken into account by the factor @xmath206 .
the kerr rotation angle is given by eq . where @xmath162 is replaced by @xmath177 .
( a ) and @xmath205 ( b ) with account for elastic scattering processes at @xmath207 , @xmath100.,title="fig : " ] ( a ) and @xmath205 ( b ) with account for elastic scattering processes at @xmath207 , @xmath100.,title="fig : " ]
in this chapter we have demonstrated the rich variety of spin - dependent phenomena in the streaming regime that can be realized in moderate electric fields in semiconductor heterostructures .
the proposed experiments , both steady - state and non - stationary , seem promising for studies of spin properties of two - dimensional electron ( or hole ) gas .
observation of photocarrier spin dynamics in the streaming regime by the pump - probe technique would allow the direct access to the time - dependent acceleration of charge carriers in the electric field . | we present results of a cross - disciplinary theoretical research at the interface of spin physics and hot - electron transport .
a moderately strong electric field is assumed to provide the streaming regime where each free charge carrier , an electron or a hole , accelerates quasiballistically in the `` passive '' region until reaching the optical - phonon energy , then emits an optical phonon and starts the next period of acceleration .
the inclusion of spin degree of freedom into the streaming - regime kinetics gives rise to rich and interesting spin - related phenomena .
firstly , in the streaming regime the spin relaxation is substantially modified , and the current - induced spin orientation is remarkably increased . under short - pulsed photoexcitation at the bottom of conduction band
the photoelectrons execute a periodic damped motion in the energy space with the period equal to the free flight time of an electron in the passive region . if the short optical pulse is circularly polarized so that the photocarriers are spin oriented , then the spin energy distribution is oscillating in time as well , which can be detected in the pump - probe time - resolved experiments .
we show that the spin - orbit splitting of the conduction band becomes a source for additional spin oscillations , periodic or aperiodic depending on the value of electric field .
* pacs * 72.25.hg , 72.25.pn , 72.25.rb , 73.63.hs .
* keywords : * spin - orbit splitting , semiconductor heterostructures , streaming |
direct detection of the progenitors of core - collapse supernovae ( ccsne ) in archival observations of nearby galaxies , particularly those from the _ hubble space telescope _ ( hst ) , has recently proved successful .
luminous red supergiant progenitors of the hydrogen rich type ii - p sne have now been identified @xcite , with initial masses above 8m@xmath4 .
however the progenitors of type ib / c sn , believed to be the cores of massive stars that have lost their outer envelopes , have so far eluded discovery ( e.g. * ? ? ?
the two nearest type ib / c sne with deep pre - explosion images ( sn2002ap in m74 at @xmath79mpc and sn2004gt in ngc4038 at @xmath714mpc ) had no progenitor detected @xcite .
the progenitors were limited to either lower mass stars that were stripped of their hydrogen - rich envelope via binary interaction , or single stars with masses @xmath8 for which radiatively driven mass loss may have been sufficient to remove the outer layers . in the case of the type ib sn 2006jc the progenitor was discovered in an lbv - like outburst state 2 years before explosion , suggesting it was a very massive star @xcite .
some fraction of type ib / c sne also produce gamma - ray bursts , hence any information on their progenitors would be critical for the field . in this letter we study the explosion site of sn 2007gr in ngc 1058 and confirm it as a type ic .
sn 2007gr was discovered by @xcite on 2007 aug 15.15ut , located at @xmath9 .
@xcite spectroscopically classified sn 2007gr as type ib / c , but suggested that the presence of strong o i @xmath107774 and ca ii ir triplet absorptions were similar to those features observed in type ic sne .
we adopt a total reddening towards sn 2007gr of @xmath11 ( valenti et al . in prep ) . where standard filter sets were employed extinctions were calculated using the laws of cardelli , clayton & mathis ( 1989 ) , and for hst observations using the @xmath12 relations of @xcite .
the pre - explosion site of sn 2007gr was observed on 2001 jul 03 with the _ wide field and planetary camera 2 _ ( wfpc2 ) on board
hst as part of programme go-9042 .
observations were made in the f450w and f814w filters ( 460s each ) and photometry was carried out using the psf - fitting photometry package hstphot @xcite .
further ground - based observations of ngc1058 were found in the 4.2-m william herschel telescope ( wht ) and the 2.5 m isaac newton telescope ( int ) archives .
the wht ingrid @xmath13band images from 2001 oct 6 had a total exposure time of 1320s and an image resolution of 12 .
the int wfc images from 2005 jan 13 were taken using @xmath14-band ( 180s ) and narrow h@xmath15 filters ( 600s ) with image quality of 09 .
ground - based adaptive optics ( ao ) observations of sn 2007gr were taken on 2007 aug 19 using altair / niri on the 8.1-m gemini ( north ) telescope .
exposures totalling 930s were acquired in @xmath6-band using the sn as the natural guide star .
the reduced image is of exceptionally high quality showing near diffraction limited resolution of 008 , with a pixel scale of 0022 providing excellent sampling of the psf .
fig.[fig : pre_and_post ] shows that this gemini image even resolves some objects which appear blended in the hst data , an effect of the 01 pixel scale of the wfpc2 chip . in order to confirm the classification of the sn ,
a spectrum was obtained on 2007 sept 11 using the wht equipped with isis ( r300b+r158r ) .
the site of sn 2007gr fell on the wf2 chip of the pre - explosion wfpc2 observations . to determine the position of the sn site accurately on these images we calculated a transformation to map the coordinate system of the post - explosion gemini image to the wfpc2 pre - explosion frames using the iraf geomap task
. positions of 19 stars common to both the wfpc2 f814w and gemini @xmath6-band images provided a general geometric transformation ( involving shifts , scales and rotation ) with an rms error of @xmath16mas .
the sn position was measured in the gemini image and transformed to the coordinate system of the f814w pre - explosion frame yielding a pixel position on the wf2 chip of [ 250.65 , 118.34 ] .
fig.[fig : pre_and_post]a & b show that the sn site is very close to , but not coincident with , an object that is bright in both the f450w and f814w images .
its position was measured as [ 250.10 , 119.56 ] with an accuracy of @xmath17mas .
the sn site is therefore 0121 west and 0056 south of this bright object .
@xcite give a position for sn2007gr on the wfpc2 images that is around 66 mas further north than our position , meaning our respective positions are not coincident within the combined errors ( combined error from this study @xmath18mas , li et al .
claim @xmath16mas ) .
we suggest that the li et al .
error has been underestimated given that they used a megacam image of the sn from the canada france hawaii telescope to perform differential astrometry , which is of much lower resolution than our gemini data .
the position of the sn on the int and wht pre - explosion observations was determined using a similar method . the int r-band image ( fig.[fig : pre_and_post]d ) shows significant flux at the sn position , but
this is almost certainly due entirely to the resolved sources observed in the hst frames .
subtraction of the r- band continuum from the @xmath19 image identifies regions of strong @xmath19 emission .
since hot ob - type stars are required to ionise the hydrogen gas , @xmath19 emission is generally associated with clusters of young , massive stars .
the image subtraction package isis 2.2 ( alard & lupton 1998 ; alard 2000 ) was used to match the psf , flux and background of the int r image to its @xmath19 counterpart .
fig.[fig : pre_and_post]e shows several regions of quite strong emission surrounding the sn site , but none exactly coincident with its position . in the wht k - band image ( fig.[fig : pre_and_post]f )
nothing is visible at or near the sn position .
sn2007gr occurred in a large , extended star - forming region that in fig.[fig : pre_and_post ] is at least 5 in diameter ( 250pc ) . within this
there are bright sources , some of which are point - like and others which are clearly extended .
most of these are likely to be compact stellar clusters rather than single individual supergiants .
compact star clusters tend to be clustered together into large star forming complexes of the order 50 - 300pc , and the recent study of the cluster complexes in m51 shows many that are morphologically similar to the region hosting sn2007gr .
hstphot photometry of the pre - explosion wfpc2 observations gave apparent magnitudes for the bright object close to the sn site of f450w @xmath20 and f814w @xmath21 .
converting these to absolute standard magnitudes using the distance to ngc 1058 and extinction corrections ( section [ sec : intro ] ) , and the colour corrections of @xcite , gave @xmath22 , and @xmath23 .
if it is a single star it is a 30 - 40m@xmath4 post main sequence supergiant star , or a more massive but extinguished ( @xmath24 mag ) main sequence object .
such bright , single stars are not without precedent in star - forming galaxies ( e.g. * ? ? ?
* ) , but they are few in number in any typical spiral due to the short lifetimes of the blue - yellow supergiant phase ( e.g. eldridge & tout 2004 )
. surrounding the sn location are several other objects of similarly high luminosity that have colours consistent with o- to f - type supergiants ( fig.[fig : cmd ] ) .
it is unlikely that one would observe such a large number of extremely massive stars during this very short evolutionary supergiant phase in such a small surface area of the disk .
suggest that sources brighter than @xmath25 are most probably clusters .
several of the objects appear extended implying that they are most probably not single stars .
most however , including the bright source nearest the sn location , are not significantly broader than the stellar psf ( [email protected] pixels ( 015 ) ) , which at a distance of 10.6mpc corresponds to 7.7pc . given its absolute magnitudes and its proximity to the sn site ( @xmath76.9pc ) one could suggest that this bright object is a coeval cluster of diameter less than @xmath77.7pc , which hosted sn 2007gr . if it could be confirmed as such a cluster
this is potentially a powerful way of determining the initial mass of the sn progenitor . in an attempt to constrain the true spatial extent of this object and others we used the _ ishape _ routine , part of the baolab package ( larsen 1999 , 2004 ) .
larsen ( 2004 ) suggests that _ ishape _ can limit the intrinsic size of a significantly detected source ( s / n @xmath750@xmath26 ) down to @xmath710 per cent of the psf fwhm , which here corresponds to @xmath71pc .
elliptical moffat profiles with a power law index of -1.5 were convolved with tinytim ( krist & hook 1997 ) wfpc2 psfs and fitted to all the objects within 3@xmath27 of the sn position .
those sources which are detected at @xmath2810@xmath26 level and have fitted fwhm values @xmath280.15 pixels ( 10 per cent of the psf fwhm ) in one or both of the hst observations are identified as extended in fig.[fig : cmd ] . the bright source nearest the sn position is best fit with a delta function , suggesting that its diameter is no larger than @xmath71pc , and that it is more likely a star rather than a cluster .
however , aperture photometry of the gemini ao image , calibrated using the two bright 2mass sources north and south of the sn , revealed that several sources around the sn location have significant k - band excesses when compared to main sequence and supergiant stellar colours ( see fig.[fig : cmd ] ) , implying that they are not individual massive stars .
two of these sources are measured as point - like by _ ishape_. unfortunately we can not measure a k - band magnitude for the object nearest the progenitor site in the gemini image due to the brilliance of the sn .
additionally the @xmath19 image ( fig.[fig : pre_and_post]e ) gives some idea of the location of the youngest ob - stars within this large star - forming region .
sources in the hst images coincident with the @xmath19 flux are marked in fig.[fig : cmd ] .
their association with significant @xmath19 emission implies that they are most likely young compact star clusters rather than single massive supergiants .
we note , however , that no significant @xmath19 emission is observed exactly coincident with the object nearest the sn site .
if this object is actually a massive single star rather than a cluster , and the progenitor is an unrelated star within this extended star - forming region , then we can attempt to derive a luminosity limit for the undetected sn progenitor .
we added fake stars of various magnitudes into the wfpc2 images at the position of the sn progenitor and tried to recover both the real and fake sources simultaneously with psf - fitting techniques .
we defined the detection limit in each filter as the magnitude of the fake source that must be added to the image in order for it to be independently detected at a position coincident with the progenitor site .
the detection limits were thereby found to be f450w @xmath29 ( @xmath30 ) and f814w @xmath31 ( @xmath32 ) .
the detection limit for the wht k - band image was found to be @xmath33 ( @xmath34 ) by adding fake stars of increasing luminosity at the sn position and determining , by eye , at what magnitude an obvious source became visible .
fig.[fig : spectra ] shows our wht + isis spectrum compared with those of well studied type ib / c sne .
the phase corresponds to about two weeks past maximum ( valenti et al . in prep . ) .
while in sn 1999dn the he i lines are prominent , they are weak ( if at all visible ) in the type ic sne and also in sn 2007gr , suggesting that sn 2007gr is of type ic .
sn 2007gr shows clear similarity to sn 1994i ( and , more marginally , to sn 2004aw ) .
however , the presence of narrower features due to metals in the spectrum of sn 2007gr makes this object rather peculiar .
if we assume that the bright object closest to the sn site is a single , massive star , one can attempt to place constraints on the properties of the sn progenitor using the detection limits derived in the previous section .
however , these magnitude limits are much shallower than those found for the progenitors of sn 2002ap @xcite and sn 2004gt @xcite .
since the sn was of type ic we can reasonably assume that its progenitor was a wolf - rayet ( w - r ) star ( e.g. eldridge & tout 2004 ) .
comparing our magnitude limit of @xmath30 with b - band photometry of w - r stars shown in fig.8 of @xcite , we see that almost all of these observed w - r stars fall below our detection limit . if , however , the bright object closest to the sn site is an unresolved stellar cluster , then it must have a diameter of less than @xmath77.7pc ( fwhm of the psf ) or perhaps less than @xmath71pc ( _ ishape _ analysis ) . in the case of the latter the progenitor
would then be on the outskirts of the cluster , although still quite possibly coeval , e.g. ngc 3603 in the milky way .
if sn 2007gr arose from one of the most evolved stars in such a coeval cluster then the cluster age would give an estimate of the age and initial mass of the progenitor .
a similar argument has been used to estimate the progenitor mass of the type ii - p sn 2004dj @xcite , through @xmath35 population fitting of the @xmath36 spectral energy distribution .
the @xmath35 code @xcite computes a coeval stellar population at a range of ages , with a user - defined initial mass function .
a cluster of mass @xmath37m@xmath4 at an age of @xmath1myrs produces the absolute @xmath38 and @xmath39 ( assuming solar metallicity , an upper mass limit of 100m@xmath4 , a salpeter imf , and a coeval burst of star - formation ) .
the geneva stellar models imply a turn - off mass of 25m@xmath4 for the cluster , while our stars stellar models ( eldridge & tout 2004 ) imply a similar mass of @xmath3m@xmath4 .
these are compatible with the minimum stellar mass thought to produce w - r stars through single stellar evolution ( e.g. * ? ? ?
* ; * ? ? ?
however this age solution is not unique as @xmath35 can produce @xmath40 when the cluster has evolved to @xmath2myrs . at this age
the turn - off mass would be between @xmath5m@xmath4 , much too low to produce a w - r star via single stellar evolution , and we would be forced to conclude that the progenitor was a lower mass w - r star resulting from an interacting binary .
this degeneracy can only be broken with the extra wavelength coverage of the host cluster .
in particular the @xmath41-band is an excellent discriminator of age , and would allow its determination to within a few myrs . at the older age
the cluster becomes much redder in the nir bands , with an estimated @xmath42 and hence we would expect to detect the source at an apparent magnitude of @xmath43 .
the detection limit of the wht k - band image lies just at this level , and this would favour the younger solution , suggesting the w - r progenitor formed from an initially 25 - 30m@xmath4 star . in approximately two years time , the sn will have faded enough that the putative cluster can be studied in detail , in the uv , optical and nir with ground - based ao imaging and hopefully a fully refurbished hst . with these higher spatial resolution and multi - wavelength images of the immediate environment
we will be able to determine if indeed it is a compact cluster and better ascertain its properties .
given the difficulties in pinning down the progenitors of ib / c sne , even with very deep pre - explosion imaging @xcite , this particular case may offer us the best opportunity to estimate the initial mass of a type ic progenitor .
the proximity and luminosity of sn 2007gr will make it one of the best studied , and modelled type ic sne to date , and it will be interesting to compare future ejecta mass estimates with our cluster turn - off mass .
observations were from : the nasa / esa hubble space telescope obtained from the data archive at the space telescope science institute ; the gemini observatory ; the isaac newton group archive from casu astronomical data centre .
we acknowledge funding through euryi scheme , esf ( sjs , rmc ) , nsf / nasa grants ast-0406740/nng04gl00 ( jrm ) , academy of finland project : 8120503 ( sm ) . | we report our attempts to locate the progenitor of the peculiar type ic sn 2007gr in _ hst _ pre - explosion images of the host galaxy , ngc 1058 . aligning adaptive optics altair / niri imaging of sn 2007gr from the gemini ( north ) telescope with the pre - explosion _ hst _ wfpc2 images , we identify the sn position on the hst frames with an accuracy of 20mas .
although nothing is detected at the sn position we show that it lies on the edge of a bright source , @xmath0mas ( 6.9pc ) from its nominal centre . based on its luminosity
we suggest that this object is possibly an unresolved , compact and coeval cluster and that the sn progenitor was a cluster member , although we note that model profile fitting favours a single bright star .
we find two solutions for the age of this assumed cluster ; @xmath1myrs and @xmath2myrs , with turn - off masses of @xmath3m@xmath4 and @xmath5m@xmath4 respectively .
pre - explosion ground - based @xmath6-band images marginally favour the younger cluster age / higher turn - off mass . assuming the sn progenitor was a cluster member , the turn - off mass provides the best estimate for its initial mass .
more detailed observations , after the sn has faded , should determine if the progenitor was indeed part of a cluster , and if so allow an age estimate to within @xmath72myrs thereby favouring either a high mass single star or lower mass interacting binary progenitor . |
forecasting changes in volatility is essential for risk management , asset pricing and scenario analysis .
indeed , models for describing and forecasting the evolution of volatility and covariance among financial assets are widely applied in industry @xcite . among the most popular approaches
are worth mentioning the multivariate extensions of garch @xcite , the stochastic covariance models @xcite and realized covariance @xcite .
however most of these econometrics tools are not able to cope with more than few assets , due to the curse of dimensionality and the increase in the number of parameters @xcite , limiting their insight into the volatility evolution to baskets of few assets only .
this is unfortunate , since gathering insights into systemic risk and the unfolding of financial crises require modelling the evolution of entire markets which are composed by large numbers of assets @xcite .
we suggest to use network filtering @xcite as a valuable tool to overcome this limitation .
correlation - based filtering networks are tools which have been widely applied to filter and reduce the complexity of covariance matrices made of large numbers of assets ( of the order of hundreds ) , representative of entire markets .
this strand of research represents an important part of the econophysics literature and has given important insights for risk management , portfolio optimization and systemic risk regulation @xcite .
the volatility of a portfolio depends on the covariance matrix of the corresponding assets @xcite .
therefore , the latter can provide insights into the former . in this work we elaborate on this connection : we show that correlation matrices can be used to predict variations of volatility , once they are analysed through the lens of network filtering .
this is quite an innovative use of correlation - based networks , which have been used mostly for descriptive analyses , with the connections with risk forecasting being mostly overlooked .
some works have shown that is possible to use dimensionality reduction techniques , such as spectral methods @xcite , as early - warning signals for systemic risk @xcite : however these approaches , although promising , do not provide proper forecasting tools , as they are affected by high false positive ratios and are not designed to predict a specific quantity . the approach we propose exploits network filtering to explicitly predict future volatility of markets made of hundreds of stocks . to this end , we introduce a new dynamical measure that quantifies the rate of change in the structure of the market correlation matrix : the `` correlation structure persistence '' @xmath0 . this quantity is derived from the structure of network filtering from past correlations . then we show how such measure exhibits significant predicting power on the market volatility , providing a tool to forecast it .
we assess the reliability of this forecasting through out - of - sample tests on two different equity datasets .
the rest of this paper is structured as follows : we first describe the two datasets we have analysed and we introduce the correlation structure persistence ; then we show how our analyses point out a strong interdependence between correlation structure persistence and future changes in the market volatility ; moreover , we describe how this result can be exploited to provide a forecasting tool useful for risk management , by presenting out - of - sample tests and false positive analysis ; then we investigate how the forecasting performance changes in time ; finally we discuss our findings and their theoretical implications .
we have analysed two different datasets of equity data . the first set ( nyse dataset )
is composed by daily closing prices of @xmath1 us stocks traded in new york stock exchange , covering 15 years from 02/01/1997 to 31/12/2012 .
the second set ( lse dataset ) is composed by daily closing prices of @xmath2 uk stocks traded in the london stock exchange , covering 13 years from 05/01/2000 to 21/08/2013 .
all stocks have been continuously traded throughout these periods of time .
these two sets of stocks have been chosen in order to provide a significant sample of the different industrial sectors in the respective markets .
for each asset @xmath3 ( @xmath4 ) we have calculated the corresponding daily log - return @xmath5 , where @xmath6 is the asset @xmath3 price at day @xmath7 .
the market return @xmath8 is defined as the average of all stocks returns : @xmath9 . in order to calculate the correlation between different assets
we have then analysed the observations by using @xmath10 moving time windows , @xmath11 with @xmath12 .
each time window contains @xmath13 observations of log - returns for each asset , totaling to @xmath14 observations .
the shift between adjacent time windows is fixed to @xmath15 trading days .
we have calculated the correlation matrix within each time window , @xmath16 , by using an exponential smoothing method @xcite that allows to assign more weight on recent observations .
the smoothing factor of this scheme has been chosen equal to @xmath17 according to previously established criteria @xcite . from each correlation matrix @xmath16
we have then computed the corresponding planar maximally filtered graph ( pmfg ) @xcite .
the pmfg is a sparse network representation of the correlation matrix that retains only a subset of most significant entries , selected through the topological criterion of being maximally planar @xcite .
such networks serve as filtering method and have been shown to provide a deep insight into the dependence structure of financial assets @xcite .
once the @xmath10 pmfgs , @xmath18 with @xmath12 , have been computed we have calculated two measures , a backward - looking and a forward - looking one .
the first is a measure that monitors the correlation structure persistence , based on a measure of pmfg similarity .
this backward - looking measure , that we call @xmath19 , relies on past data only and indicates how slowly the correlation structure measured at time window @xmath11 is differentiating from structures associated to previous time windows .
the forward - looking measure is the volatility ratio @xmath20 @xcite , that at each time window quantifies how good the market volatility measured at @xmath11 is as a proxy for the next time - window volatility . unlike @xmath19 ,
the value of @xmath20 is not known at the end of @xmath11 .
[ fig : time_windows ] shows a graphical representation of the time window set - up . in the following
we define the two measures : [ sec : methods ] * * correlation structure persistence @xmath19 * : we define the correlation structure persistence at time @xmath11 as : + @xmath21 + where @xmath22 is an exponential smoothing factor , @xmath23 is a parameter and @xmath24 is the fraction of edges in common between the two pmfgs @xmath18 and @xmath25 , called `` edge survival ratio '' @xcite . in formula , @xmath24 reads : + @xmath26 + where @xmath27 is the number of edges ( links ) in the two pmfgs ( constant and equal to @xmath28 for a pmfg @xcite ) , and @xmath29 ( @xmath30 ) represents the edge - sets of pmfg at @xmath31 ( @xmath32 ) .
the correlation structure persistence @xmath19 is therefore a weighted average of the similarity ( as measured by the edge survival ratio ) between @xmath18 and the first @xmath23 previous pmfgs , with an exponential smoothing scheme that gives more weight to those pmfgs that are closer to @xmath11 .
the parameter @xmath33 in eq .
[ eq : es ] can be calculated by imposing @xmath34 .
intuitively , @xmath19 measures how slowly the change of correlation structure is occurring in the near past of @xmath11 . *
* volatility ratio @xmath20 * @xcite : in order to quantify the agreement between the estimated and the realized risk we here make use of the volatility ratio , a measure which has been used @xcite for this purpose and defined as follows : + @xmath35 + where @xmath36 is the realized volatility of the average market return @xmath8 computed on the time window @xmath37 ; @xmath38 is the estimated volatility of @xmath8 computed on time window @xmath11 , by using the same exponential smoothing scheme @xcite described for the correlation @xmath16 .
specifically , @xmath37 is the time window of length @xmath39 that follows immediately @xmath11 : if @xmath40 is the last observation in @xmath11 , @xmath37 covers observations from @xmath41 to @xmath42 ( fig .
[ fig : time_windows ] ) .
therefore the ratio in eq .
[ eq : q ] estimates the agreement between the market volatility estimated with observations in @xmath11 and the actual market volatility observed over an investment in the @xmath43 assets over @xmath37 .
if @xmath44 , then the historical data gathered at @xmath11 has underestimated the ( future ) realized volatilty , whereas @xmath45 indicates overestimation .
+ let us stress that @xmath20 provides an information on the reliability of the covariance estimation too , given the relation between market return volatility and covariance @xcite : + @xmath46 + @xmath47 + where @xmath48 and @xmath49 are respectively the estimated and realized covariances . to investigate the relation between @xmath19 and @xmath20 we have calculated the two quantities with different values of @xmath13 and @xmath23 in eqs .
[ eq : es ] and [ eq : q ] , to assess the robustness against these parameters .
specifically , we have used @xmath50 trading days , that correspond to time windows of length 1 , 2 , 3 and 4 years respectively ; @xmath51 , that correspond ( given @xmath15 trading days ) to an average in eq .
[ eq : es ] reaching back to @xmath52 , @xmath53 , @xmath54 and @xmath55 trading days respectively .
@xmath39 has been chosen equal to @xmath54 trading days ( one year ) for all the analysis . in fig .
[ fig : es_matrices ] we show the @xmath56 matrices ( eq . [ eq : es_ab ] ) for the nyse and lse dataset , for @xmath57 .
we can observe a block structure with periods of high structural persistence and other periods whose correlation structure is changing faster .
in particular two main blocks of high persistence can be found before and after the 2007 - 2008 financial crisis ; a similar result was found in a previous work @xcite with a different measure of similarity .
these results are confirmed for all values of @xmath13 considered . in fig .
[ fig : plot_342 ] we show @xmath19 and @xmath20 as a function of time , for @xmath57 and @xmath58 .
as expected , main peaks of @xmath20 occur during the months before the most turbulent periods in the stock market , namely the 2002 market downturn and the 2007 - 08 credit crisis .
interestingly , the corresponding @xmath19 seems to follow a specular trend .
this is confirmed by explicit calculation of pearson correlation between the two signals , reported in tabs .
[ tab : corr_4 ] - [ tab : corr_8 ] : as one can see , for all combinations of parameters the correlation is negative .
* @xmath24 matrices for @xmath59 , for nyse ( left ) and lse dataset ( right)*. a block - like structure can be observed in both datasets , with periods of high structural persistence and other periods whose correlation structure is changing faster .
the 2007 - 2008 financial crisis marks a transition between two main blocks of high structural persistence.,title="fig : " ] * @xmath24 matrices for @xmath59 , for nyse ( left ) and lse dataset ( right)*. a block - like structure can be observed in both datasets , with periods of high structural persistence and other periods whose correlation structure is changing faster .
the 2007 - 2008 financial crisis marks a transition between two main blocks of high structural persistence.,title="fig : " ] * @xmath19 and @xmath20 signals represented for @xmath57 and @xmath58 * , for both nyse ( left graph ) and lse ( right graph ) datasets .
it is evident the anticorrelation between the two signals .
the financial crisis triggers a major drop in the structural persistence and a corresponding peak in @xmath20 .
, title="fig : " ] * @xmath19 and @xmath20 signals represented for @xmath57 and @xmath58 * , for both nyse ( left graph ) and lse ( right graph ) datasets .
it is evident the anticorrelation between the two signals .
the financial crisis triggers a major drop in the structural persistence and a corresponding peak in @xmath20 .
, title="fig : " ] * @xmath60 matrices for @xmath59 , for nyse ( left ) and lse dataset ( right)*. a block - like structure can be observed in both datasets , with periods of high structural persistence and other periods whose correlation structure is changing faster .
the blocks of high similarity show higher compactness than in fig .
[ fig : es_matrices].,title="fig : " ] * @xmath60 matrices for @xmath59 , for nyse ( left ) and lse dataset ( right)*. a block - like structure can be observed in both datasets , with periods of high structural persistence and other periods whose correlation structure is changing faster .
the blocks of high similarity show higher compactness than in fig .
[ fig : es_matrices].,title="fig : " ] in order to check the significance of this anticorrelation we can not rely on standard tests on pearson coefficient , such as fisher transform @xcite , as they assume i.i.d .
series @xcite .
our time series are instead strongly autocorrelated , due to the overlapping between adjacent time windows . therefore we have calculated confidence intervals by performing a block bootstrapping test @xcite .
this is a variation of the bootstrapping test @xcite , conceived to take into account the autocorrelation structure of the bootstrapped series .
the only free parameter in this method is the block length , that we have chosen applying the optimal selection criterion proposed in literature @xcite : such criterion is adaptive on the autocorrelation strength of the series as measured by the correlogram .
we have found , depending on the parameters @xmath13 and @xmath23 , optimal block lengths ranging from 29 to 37 , with a mean of 34 ( corresponding to 170 trading days ) . by performing block bootstrapping tests we have therefore estimated confidence intervals for the true correlation between @xmath19 and @xmath20 ; in tabs .
[ tab : corr_4 ] - [ tab : corr_8 ] correlations whose @xmath61 and @xmath62 confidence intervals ( ci ) do not include zero are marked with one and two stars respectively .
as we can see , 14 out of 16 correlation coefficients are significantly different from zero within @xmath61 ci in the nyse dataset , and 12 out of 16 in the lse dataset . for what concerns the @xmath62 ci
, we observe 13 out of 16 for the nyse and 9 out of 16 for the lse dataset .
non - significant correlations appear only for @xmath63 , suggesting that this length is too small to provide a reliable measure of structural persistence .
very similar results are obtained by using minimum spanning tree ( mst ) @xcite instead of pmfg as network filtering .
given the interpretation of @xmath19 and @xmath20 given above , anticorrelation implies that an increase in the `` speed '' of correlation structure evolution ( low @xmath19 ) is likely to correspond to underestimation of future market volatility from historical data ( high @xmath64 ) ) , whereas when the structure evolution `` slows down '' ( high @xmath19 ) there is indication that historical data is likely to provide an overestimation of future volatility .
this means that we can use @xmath19 as a valuable predictor of current historical data reliability .
this result is to some extent surprising as @xmath19 is derived from pmfgs topology , which in turns depends only on the ranking of correlations and not on their actual value : yet , this information provides meaningful information about the future market volatility and therefore about the future covariance . in principle other measures of correlation ranking structure , more straightforward than the correlation persistence @xmath19 , might capture the same interplay with @xmath20 .
we have therefore considered also the metacorrelation @xmath65 , that is the pearson correlation computed between the coefficients of correlation matrices at @xmath11 and @xmath66 ( see methods for more details ) .
such measure does not make use of pmfg .
[ fig : meta_matrices ] displays the similarity matrices obtained with this measure for nyse and lse datasets : we can observe again block - like structures , that however carry different information from the @xmath56 in fig .
[ fig : es_matrices ] ; in particular , blocks show higher intra - similarity and less structure . similarly to eq .
[ eq : es ] , we have then defined @xmath67 as the weighted average over @xmath23 past time windows ( see methods ) . in tabs . [
tab : corr_meta_nyse ] and [ tab : corr_meta_lse ] we show the correlation between @xmath67 and @xmath20 . as we can see , although an anticorrelation is present for each combination of parameters @xmath13 and @xmath23 , correlation coefficients are systematically closer to zero than in tabs .
[ tab : corr_4 ] - [ tab : corr_8 ] , where @xmath19 was used .
moreover the number of significant pearson coefficients , according to the block bootstrapping , decreases to 12 out of 16 in nyse and to 10 out of 16 in lse dataset .
since @xmath68 does not make use of pmfg , this result suggests that the filtering procedure associated to correlation - based networks is a necessary step for capturing at best the correlation ranking evolution and its interplay with the volatility ratio .
cc | c c c c | & & + & & * 10 * & * 25 * & * 50 * & * 100 * + & & -0.2129 & -0.2224 & @xmath69 & @xmath70 + & & @xmath71 & @xmath72 & @xmath73 & @xmath74 + & & @xmath75 & @xmath76 & @xmath77 & @xmath78 + & & @xmath79 & @xmath80 & @xmath81 & @xmath82 + cc | c c c c | & & + & & * 10 * & * 25 * & * 50 * & * 100 * + & & @xmath83 & @xmath84 & -0.1872 & @xmath85 + & & @xmath86 & @xmath87 & @xmath88 & @xmath89 + & & @xmath90 & @xmath91 & @xmath92 & @xmath93 + & & @xmath94 & @xmath95 & @xmath96 & @xmath97 + cc | c c c c | & & + & & * 10 * & * 25 * & * 50 * & * 100 * + & & -0.0992 & -0.0754 & -0.1055 & -0.1157 + & & -0.2146 & -0.2232 & -0.2309 & -0.2753 + & & -0.2997 & @xmath98 & @xmath99 & @xmath100 + & & @xmath101 & @xmath102 & @xmath103 & @xmath104 + cc | c c c c | & & + & & * 10 * & * 25 * & * 50 * & * 100 * + & & -0.1470 & -0.1095 & -0.1326 & -0.1720 + & & @xmath105 & -0.2113 & @xmath106 & @xmath107 + & & @xmath108 & @xmath109 & @xmath110 & @xmath111 + & & @xmath112 & -0.2954 & -0.3163 & @xmath113 + * partition of data into training ( left graphs ) and test ( right graphs ) set*. training sets are used to regress @xmath114 against @xmath19 , in order to estimate the coefficents in the logistic regression and therefore identify the regression threshold , shown as a vertical continuous line . the test sets are used to test the forecasting performance of such regression on a subset of data that has not been used for regression ; the model predicts @xmath115 ( @xmath44 ) if @xmath19 is greater than the regression threshold , and @xmath116 ( @xmath45 ) otherwise.,title="fig : " ] * partition of data into training ( left graphs ) and test ( right graphs ) set*. training sets are used to regress @xmath114 against @xmath19 , in order to estimate the coefficents in the logistic regression and therefore identify the regression threshold , shown as a vertical continuous line . the test sets are used to test the forecasting performance of such regression on a subset of data that has not been used for regression ; the model predicts @xmath115 ( @xmath44 ) if @xmath19 is greater than the regression threshold , and @xmath116 ( @xmath45 ) otherwise.,title="fig : " ] * partition of data into training ( left graphs ) and test ( right graphs ) set*. training sets are used to regress @xmath114 against @xmath19 , in order to estimate the coefficents in the logistic regression and therefore identify the regression threshold , shown as a vertical continuous line . the test sets are used to test the forecasting performance of such regression on a subset of data that has not been used for regression ; the model predicts @xmath115 ( @xmath44 ) if @xmath19 is greater than the regression threshold , and @xmath116 ( @xmath45 ) otherwise.,title="fig : " ] * partition of data into training ( left graphs ) and test ( right graphs ) set*. training sets are used to regress @xmath114 against @xmath19 , in order to estimate the coefficents in the logistic regression and therefore identify the regression threshold , shown as a vertical continuous line .
the test sets are used to test the forecasting performance of such regression on a subset of data that has not been used for regression ; the model predicts @xmath115 ( @xmath44 ) if @xmath19 is greater than the regression threshold , and @xmath116 ( @xmath45 ) otherwise.,title="fig : " ] in this section we evaluate how well the correlation structure persistence @xmath19 can forecast the future through its relation with the forward - looking volatility ratio @xmath20 .
in particular we focus on estimating whether @xmath20 is greater or less than @xmath117 : this information , although less complete than a precise estimation of @xmath20 , gives us an important insight into possible overestimation ( @xmath45 ) or underestimation ( @xmath44 ) of future volatility .
we have proceeded as follows .
given a choice of parameters @xmath13 and @xmath23 , we have calculated the corresponding set of pairs @xmath118 , with @xmath12 .
then we have defined @xmath114 as the categorical variable that is @xmath119 if @xmath45 and @xmath117 if @xmath44 .
finally we have performed a logistic regression of @xmath114 against @xmath19 : namely , we assume that @xcite : @xmath120 where @xmath121 is the sigmoid function @xmath122 @xcite ; we estimate parameters @xmath123 and @xmath124 from the observations @xmath125 through maximum likelihood @xcite . once the model has been calibrated , given a new observation @xmath126 we have predicted @xmath127 if @xmath128 , and @xmath129 otherwise . this classification criterion , in a case with only one predictor , corresponds to classify @xmath130 according to whether @xmath131 is greater or less than a threshold @xmath132 which depends on @xmath123 and @xmath124 , as shown in fig .
[ fig : training_test ] ( right graphs ) for a particular choice of parameters .
therefore the problem of predicting whether market volatility will increase or decrease boils down to a classification problem @xcite with @xmath19 as predictor and @xmath114 as target variable .
we have made use of a logistic regression because it is more suitable than a polynomial model for dealing with classification problems @xcite .
other classification algorithms are available ; we have chosen the logistic regression due to its simplicity .
we have also implemented the knn algorithm @xcite and we have found that it provides similar outcomes but worse results in terms of the forecasting performance metrics that we discuss in this section .
we have then evaluated the goodness of the logistic regression at estimating @xmath130 given a new observation @xmath131 . to this end , we have computed three standard metrics for assessing the performance of a classification method : the probability of successful forecasting @xmath133 , the true positive rate @xmath134 and the false positive rate @xmath135 . @xmath133 represents the expected fraction of correct predictions , @xmath134 is the method goodness at identifying true positives ( in this case , actual increases in volatility ) and @xmath135 quantifies the method tendency to false positives ( predictions of volatility increase when the volatility will actually decrease ) : see methods for more details .
overall these metrics provide a complete summary of the model goodness at predicting changes in the market volatility @xcite . in order to avoid overfitting
we have estimated the metrics above by means of an out - of - sample procedure @xcite .
we have divided our dataset into two periods , a training set and a test set . in the training
set we have calibrated the logistic equation in eq .
[ eq : logistic_regression ] , estimating the parameters @xmath123 and @xmath124 ; in the test set we have used the calibrated model to measure the goodness of the model predictions by computing the measures of performance in eq .
[ eq : prob_forecast]-[eq : fpr ] . in fig .
[ fig : training_test ] this division is shown for a particular choice of @xmath13 and @xmath23 , for both nyse and lse dataset . in this example
the percentage of data included in the test set ( let us call it @xmath136 ) is @xmath137 .
probabilities of successful forecasting @xmath133 are reported in tabs . [
tab : nyse_p_outs ] and [ tab : lse_p_outs ] , for @xmath138 .
as we can see @xmath133 is higher than @xmath139 for all combinations of parameters in nyse dataset , and in almost all combinations for lse dataset .
stars mark those values of @xmath133 that are significantly higher than the same probability obtained by using the most recent value of @xmath140 instead of @xmath19 as a predictor for @xmath20 in the logistic regression ( let us call @xmath141 such probability ) .
specifically , we have defined a null model where variations from such probability @xmath141 are due to random fluctuations only ; given @xmath10 observations , such fluctuations follow a binomial distribution @xmath142 , with mean @xmath143 and variance @xmath144 . then p - values have been calculated by using this null distribution for each combination of parameters .
this null hypothesis accounts for the predictability of @xmath20 that is due to the autocorrelation of @xmath20 only ; therefore @xmath133 significantly higher than the value expected under this hypothesis implies a forecasting power of @xmath19 that is not explained by the autocorrelation of @xmath20 . from the table
we can see that @xmath133 is significant in 12 out of 16 combinations of parameters for nyse dataset , and in 13 out of 16 for lse dataset .
this means that correlation persistence is a valuable predictor for future average correlation , able to outperform forecasting method based on past average correlation trends .
these results are robust against changes of @xmath136 , as long as the training set is large enough to allow an accurate calibration of the logistic regression .
we have found this condition is satisfied for @xmath145 .
however @xmath133 does not give any information on the method ability to distinguish between true and false positives . to investigate this aspect we need @xmath134 and @xmath135 .
a traditional way of representing both measures from a binary classifier is the so - called `` receiver operating characteristic '' ( roc ) curve @xcite . in a roc plot , @xmath134
is plotted against @xmath135 as the discriminant threshold is varied .
the discriminant threshold @xmath146 is the value of the probability in eq .
[ eq : logistic_regression ] over which we classify @xmath115 : the higher @xmath146 is , the less likely the method is to classify @xmath115 ( in the analysis on @xmath133 we chose @xmath147 ) .
ideally , a perfect classifier would yield @xmath148 for all @xmath149 , whereas a random classifier is expected to lie on the line @xmath150 .
therefore a roc curve which lies above the line @xmath150 indicates a classifier that is better than chance at distinguishing true from false positives @xcite . as one can see from fig .
[ fig : roc_curve ] , the roc curve s position depends on the choice of parameters @xmath13 and @xmath23 . in this respect
our classifier performs better for low values of @xmath23 and @xmath13 .
this can be quantified by measuring the area under the roc curve ; such measure , often denoted by auc @xcite , is shown in tabs .
[ tab : auc_nyse_outs]-[tab : auc_lse_outs ] . for both datasets
the optimal choice of parameters is @xmath151 and @xmath152 .
cc | c c c c | & & + & & * 10 * & * 25 * & * 50 * & * 100 * + & & 0.546 & 0.560 * & 0.599 * * & 0.539 * * + & & 0.704 * * & 0.695 * * & 0.658 * * & 0.605 * * + & & 0.634 * & 0.585 & 0.539 & 0.708 * + & & 0.704 * & 0.7638 * * & 0.839 * * & 0.860 + cc | c c c c | & & + & & * 10 * & * 25 * & * 50 * & * 100 * + & & 0.616 * * & 0.645 * * & 0.612 * * & 0.568 * * + & & 0.652 * * & 0.635 * * & 0.598 * * & 0.393 + & & 0.651 * * & 0.560 * * & 0.453 * * & 0.412 + & & 0.544 * * & 0.573 * * & 0.706 * * & 0.689 + * receiver operating characteristic ( roc ) curve*. upper graph : true positive rate ( tpr ) against false positive rate ( fpr ) as the discriminant threshold @xmath146 of the classifier is varied , for each combination of parameters @xmath13 and @xmath23 in the nyse dataset
. the closer the curve is to the upper left corner of each graph , the better is the classifier compared to chance .
bottom graph : true positive rate ( tpr ) against false positive rate ( fpr ) as the discriminant threshold @xmath146 of the classifier is varied , for each combination of parameters @xmath13 and @xmath23 in the lse dataset.,title="fig : " ] * receiver operating characteristic ( roc ) curve*. upper graph : true positive rate ( tpr ) against false positive rate ( fpr ) as the discriminant threshold @xmath146 of the classifier is varied , for each combination of parameters @xmath13 and @xmath23 in the nyse dataset .
the closer the curve is to the upper left corner of each graph , the better is the classifier compared to chance .
bottom graph : true positive rate ( tpr ) against false positive rate ( fpr ) as the discriminant threshold @xmath146 of the classifier is varied , for each combination of parameters @xmath13 and @xmath23 in the lse dataset.,title="fig : " ] .[tab : auc_nyse_outs ] nyse dataset : * area under the curve ( auc ) * , measured from the roc curve in fig .
[ fig : roc_curve ] .
values greater than 0.5 indicate that the classifier performs better than chance . [ cols="^,^,^,^,^,^ " , ] * fraction of successful predictions as a function of time*. nyse ( left graph ) and lse dataset ( right graph ) .
forecasting is based on logistic regression with predictor @xmath153 ( top graphs ) and most recent value of @xmath20 ( bottom graphs ) .
horizontal lines represent the average over the entire period.,title="fig : " ] * fraction of successful predictions as a function of time*. nyse ( left graph ) and lse dataset ( right graph ) .
forecasting is based on logistic regression with predictor @xmath153 ( top graphs ) and most recent value of @xmath20 ( bottom graphs ) .
horizontal lines represent the average over the entire period.,title="fig : " ] in this section we look at how the forecasting performance changes at different time periods . in order to explore this aspect we have counted at each time window @xmath11 the number @xmath154 of @xmath114 predictions ( out of the 16 predictions corresponding to as many combinations of @xmath13 and @xmath23 ) that have turned out to be correct
; we have then calculated the fraction of successful predictions @xmath155 as @xmath156 . in this way
@xmath155 is a proxy for the goodness of our method at each time window .
logistic regression parameters @xmath123 and @xmath124 have been calibrated by using the entire time period as training set , therefore this amounts to an in - sample analysis . in fig .
[ fig : forecast_time ] we show the fraction of successful predictions for both nyse and lse datasets ( upper graphs , blue circles ) . for comparison
we also show the same measure obtained by using the most recent value of @xmath20 as predictor ( bottom graphs ) ; as in the previous section , it represents a null model that makes prediction by using only the past evolution of @xmath20 . as we can see , both predictions based on @xmath19 and on past values of @xmath20 display performances changing in time . in particular
@xmath155 drops just ahead of the main financial crises ( the market downturn in march 2002 , 2007 - 2008 financial crisis , euro zone crisis in 2011 ) ; this is probably due to the abrupt increase in volatility that occurred during these events and that the models took time to detect .
after these drops though performances based on @xmath19 recover much more rapidly than those based on past value of @xmath20 .
for instance in the first months of 2007 our method shows quite high @xmath155 ( more than @xmath157 of successful predictions ) , being able to predict the sharp increase in volatility to come in 2008 while predictions based on @xmath20 fail systematically until 2009 .
overall , predictions based on correlation structure persistence appear to be more reliable ( as shown by the average @xmath155 over all time windows , the horizontal lines in the plot ) and faster at detecting changes in market volatility .
in this paper we have proposed a new tool for forecasting market volatility based on correlation - based information filtering networks and logistic regression , useful for risk and portfolio management .
the advantage of our approach over traditional econometrics tools , such as multivariate garch and stochastic covariance models , is the `` top - down '' methodology that treats correlation matrices as the fundamental objects , allowing to deal with many assets simultaneously ; in this way the curse of dimensionality , that prevents e.g. multivariate garch to deal with more than few assets , is overcome .
we have proven the forecasting power of this tool by means of out - of - sample analyses on two different stock markets ; the forecasting performance has been proven to be statistically significant against a null model , outperforming predictions based on past market correlation trends . moreover we have measured the roc curve and identified an optimal region of the parameters in terms of true positive and false positive trade - off .
the temporal analysis indicates that our method is able to adapt to abrupt changes in the market , such as financial crises , more rapidly than methods based on past volatility .
this forecasting tool relies on an empirical fact that we have reported in this paper for the first time .
specifically , we have shown that there is a deep interplay between market volatility and the rate of change of the correlation structure .
the statistical significance of this relation has been assessed by means of a block - bootstrapping technique .
an analysis based on metacorrelation has revealed that this interplay is better highlighted when filtering based on planar maximally filtered graphs is used to estimate the correlation structure persistence .
this finding sheds new light into the dynamic of correlation .
the topology of planar maximally filtered graphs depends on the ranking of the @xmath158 pairs of cross - correlations ; therefore an increase in the rate of change in pmfgs topology points out a faster change of this ranking .
our result indicates that such increase is typically followed by a rise in the market volatility , whereas decreases are followed by drops .
a possible interpretation of this is related to the dynamics of risk factors in the market .
indeed higher volatility in the market is associated to the emergence of a ( possibly new ) risk factor that makes the whole system unstable ; such transition could be anticipated by a quicker change of the correlation ranking , triggered by the still emerging factor and revealed by the correlation structure persistence .
such persistence can therefore be a powerful tool for monitoring the emergence of new risks , valuable for a wide range of applications , from portfolio management to systemic risk regulation .
moreover this interpretation would open interesting connections with those approaches to systemic risk that make use of principal component analysis , monitoring the emergence of new risk factors by means of spectral methods @xcite .
we plan to investigate all these aspects in a future work .
given two correlation matrices @xmath159 and @xmath160 at two different time windows @xmath11 and @xmath66 , their metacorrelation @xmath65 is defined as follows : @xmath161[\langle \rho^2_{ij}(t_b ) \rangle_{ij } - \langle \rho_{ij}(t_b ) \rangle_{ij}^2 ] } } , \label{eq : z}\ ] ] where @xmath162 is the average over all couples of stocks @xmath163 .
similarly to eq .
[ eq : es ] we have then defined @xmath67 as the weighted average over @xmath23 past time windows : @xmath164 with reference to figs .
[ fig : training_test ] b ) and d ) , let us define the number of observations in each quadrant @xmath165 ( @xmath166 ) as @xmath167 . in the terminology of classification techniques @xcite , @xmath168 is the number of true positive ( observations for which the model correctly predicted @xmath169 ) , @xmath170 is the number of true negative ( observations for which the model correctly predicted @xmath171 ) , @xmath172 the number of false negative ( observations for which the model incorrectly predicted @xmath171 ) and @xmath173 the number of false positive ( observations for which the model incorrectly predicted @xmath115 ) .
we have then computed the following measures of quality of classification , that are the standard metrics for assessing the performances of a classification method @xcite : * * probability of successful forecasting ( @xmath133 ) * @xcite : represents the method probability of a correct prediction , expressed as fraction of observed @xmath19 values through which the method has successfully identified the correspondent value of @xmath174 . in classification problems ,
sometimes , the error rate @xmath175 is used @xcite , which is simply @xmath176 .
@xmath133 is computed as follows : + @xmath177 * * true positive rate ( @xmath134 ) * @xcite : it is the probability of predicting @xmath115 , conditional to the fact that the real @xmath114 is indeed @xmath117 ( that is , to predict an increase in volatility when the volatility will indeed increase ) ; it represents the method sensitivity to increase in volatility .
it is also called `` recall '' @xcite . in formula : + @xmath178 * * false positive rate ( @xmath135 ) * @xcite : it is the probability of predicting @xmath115 , conditional to the fact that the real @xmath114 is instead @xmath119 ( that is , to predict an increase in volatility when the volatility will actually decrease ) .
it is also called `` 1-specificity '' @xcite . in formula : + @xmath179 10 url # 1`#1`urlprefix[2]#2 [ 2][]#2 _ _ ( , ) . & . _ _ ( ) . & _ _ ( , ) . , , , & . _ _ * * , ( ) . , & . _ _ * * , ( ) . .
_ _ * * , ( ) . , , & .
_ _ * * , ( ) . .
_ _ * * , ( ) . , , & .
_ _ * * , ( ) . & .
_ _ * * , ( ) . , , , & .
_ _ * * , ( ) . , & .
_ _ * * , ( ) . , & .
_ _ * * , ( ) .
_ _ ( , ) .
_ _ * * , ( ) .
_ _ * * , ( ) . , , & .
_ _ * * , ( ) . , & .
_ _ * * , ( ) . , & .
_ _ * * , ( ) . , & .
_ _ * * , ( ) . .
_ _ * * , ( ) .
_ et al . _ . _
_ * * , ( ) .
_ _ * * , ( ) . , , & . _ _ ( ) . , &
_ _ * * , ( ) . , &
_ _ * * , ( ) . , & .
_ _ * * , ( ) . & .
_ _ * * , ( ) . & .
_ _ * * , ( ) . , & .
_ _ ( ) .
. _ _ * * , ( ) . & _ _ ( , ) . .
_ _ * * , ( ) . .
_ _ * * , ( ) .
_ _ * * , ( ) .
_ _ * * , ( ) .
, , & _ _ ( , ) .
_ _ ( , ) .
_ _ ( , ) . .
the authors wish to thank bloomberg for providing the data .
tdm wishes to thank the cost action td1210 for partially supporting this work .
ta acknowledges support of the uk economic and social research council ( esrc ) in funding the systemic risk centre ( es / k002309/1 ) .
* competing financial interests . *
the authors declare no competing financial interest . | we discovered that past changes in the market correlation structure are significantly related with future changes in the market volatility . by using correlation - based information filtering networks we device a new tool for forecasting the market volatility changes .
in particular , we introduce a new measure , the `` correlation structure persistence '' , that quantifies the rate of change of the market dependence structure .
this measure shows a deep interplay with changes in volatility and we demonstrate it can anticipate market risk variations .
notably , our method overcomes the curse of dimensionality that limits the applicability of traditional econometric tools to portfolios made of a large number of assets .
we report on forecasting performances and statistical significance of this tool for two different equity datasets .
we also identify an optimal region of parameters in terms of true positive and false positive trade - off , through a roc curve analysis .
we find that our forecasting method is robust and it outperforms predictors based on past volatility only .
moreover the temporal analysis indicates that our method is able to adapt to abrupt changes in the market , such as financial crises , more rapidly than methods based on past volatility . |
Subsets and Splits
No saved queries yet
Save your SQL queries to embed, download, and access them later. Queries will appear here once saved.